From 5a72c1af0ae24bcc7c535b742bef8f17e3ffca7c Mon Sep 17 00:00:00 2001 From: lunrenyi Date: Thu, 2 Apr 2026 15:29:08 +0800 Subject: [PATCH 1/6] chore: remove deprecated x-cmd skill documentation --- .../minimax-multimodal-toolkit/SKILL.md | 88 ++++++++------ .../scripts/image/generate_image.sh | 114 +++++++++++++----- data/x-cmd/aider/SKILL.md | 39 ------ data/x-cmd/aider/SKILL.zh.md | 39 ------ data/x-cmd/arp/SKILL.md | 45 ------- data/x-cmd/arp/SKILL.zh.md | 45 ------- data/x-cmd/asdf/SKILL.md | 46 ------- data/x-cmd/asdf/SKILL.zh.md | 46 ------- data/x-cmd/assert/SKILL.md | 55 --------- data/x-cmd/assert/SKILL.zh.md | 55 --------- data/x-cmd/brew/SKILL.md | 47 -------- data/x-cmd/brew/SKILL.zh.md | 47 -------- data/x-cmd/bun/SKILL.md | 34 ------ data/x-cmd/bun/SKILL.zh.md | 34 ------ data/x-cmd/buse/SKILL.md | 51 -------- data/x-cmd/buse/SKILL.zh.md | 51 -------- data/x-cmd/bwh/SKILL.md | 55 --------- data/x-cmd/bwh/SKILL.zh.md | 55 --------- data/x-cmd/cal/SKILL.md | 36 ------ data/x-cmd/cal/SKILL.zh.md | 36 ------ data/x-cmd/cb/SKILL.md | 26 ---- data/x-cmd/cb/SKILL.zh.md | 26 ---- data/x-cmd/ccal/SKILL.md | 45 ------- data/x-cmd/ccal/SKILL.zh.md | 45 ------- data/x-cmd/ccmd/SKILL.md | 46 ------- data/x-cmd/ccmd/SKILL.zh.md | 46 ------- data/x-cmd/cd/SKILL.md | 52 -------- data/x-cmd/cd/SKILL.zh.md | 52 -------- data/x-cmd/cht/SKILL.md | 46 ------- data/x-cmd/cht/SKILL.zh.md | 46 ------- data/x-cmd/claude/SKILL.md | 69 ----------- data/x-cmd/claude/SKILL.zh.md | 60 --------- data/x-cmd/codex/SKILL.md | 63 ---------- data/x-cmd/codex/SKILL.zh.md | 60 --------- data/x-cmd/coincap/SKILL.md | 46 ------- data/x-cmd/coincap/SKILL.zh.md | 46 ------- data/x-cmd/cpu/SKILL.md | 36 ------ data/x-cmd/cpu/SKILL.zh.md | 36 ------ data/x-cmd/crush/SKILL.md | 51 -------- data/x-cmd/crush/SKILL.zh.md | 51 -------- data/x-cmd/ddgo/SKILL.md | 47 -------- data/x-cmd/ddgo/SKILL.zh.md | 47 -------- data/x-cmd/deepseek/SKILL.md | 49 -------- data/x-cmd/deepseek/SKILL.zh.md | 49 -------- data/x-cmd/deno/SKILL.md | 28 ----- data/x-cmd/deno/SKILL.zh.md | 28 ----- data/x-cmd/dingtalk/SKILL.md | 39 ------ data/x-cmd/dingtalk/SKILL.zh.md | 39 ------ data/x-cmd/discord/SKILL.md | 31 ----- data/x-cmd/discord/SKILL.zh.md | 31 ----- data/x-cmd/dns/SKILL.md | 44 ------- data/x-cmd/dns/SKILL.zh.md | 44 ------- data/x-cmd/docker/SKILL.md | 47 -------- data/x-cmd/docker/SKILL.zh.md | 47 -------- data/x-cmd/doubao/SKILL.md | 39 ------ data/x-cmd/doubao/SKILL.zh.md | 39 ------ data/x-cmd/elv/SKILL.md | 38 ------ data/x-cmd/elv/SKILL.zh.md | 38 ------ data/x-cmd/facl/SKILL.md | 33 ----- data/x-cmd/facl/SKILL.zh.md | 33 ----- data/x-cmd/feishu/SKILL.md | 38 ------ data/x-cmd/feishu/SKILL.zh.md | 38 ------ data/x-cmd/ffmpeg/SKILL.md | 47 -------- data/x-cmd/ffmpeg/SKILL.zh.md | 47 -------- data/x-cmd/find/SKILL.md | 46 ------- data/x-cmd/find/SKILL.zh.md | 46 ------- data/x-cmd/fish/SKILL.md | 38 ------ data/x-cmd/fish/SKILL.zh.md | 38 ------ data/x-cmd/fjo/SKILL.md | 26 ---- data/x-cmd/fjo/SKILL.zh.md | 26 ---- data/x-cmd/free/SKILL.md | 47 -------- data/x-cmd/free/SKILL.zh.md | 47 -------- data/x-cmd/fsiter/SKILL.md | 57 --------- data/x-cmd/fsiter/SKILL.zh.md | 57 --------- data/x-cmd/gcal/SKILL.md | 44 ------- data/x-cmd/gcal/SKILL.zh.md | 44 ------- data/x-cmd/gddy/SKILL.md | 55 --------- data/x-cmd/gddy/SKILL.zh.md | 55 --------- data/x-cmd/gemini/SKILL.md | 54 --------- data/x-cmd/gemini/SKILL.zh.md | 54 --------- data/x-cmd/gg/SKILL.md | 48 -------- data/x-cmd/gg/SKILL.zh.md | 48 -------- data/x-cmd/gh/SKILL.md | 48 -------- data/x-cmd/gh/SKILL.zh.md | 48 -------- data/x-cmd/gitconfig/SKILL.md | 31 ----- data/x-cmd/gitconfig/SKILL.zh.md | 31 ----- data/x-cmd/githook/SKILL.md | 27 ----- data/x-cmd/githook/SKILL.zh.md | 27 ----- data/x-cmd/gl/SKILL.md | 38 ------ data/x-cmd/gl/SKILL.zh.md | 38 ------ data/x-cmd/gm/SKILL.md | 45 ------- data/x-cmd/gm/SKILL.zh.md | 45 ------- data/x-cmd/gram/SKILL.md | 61 ---------- data/x-cmd/gram/SKILL.zh.md | 56 --------- data/x-cmd/grep/SKILL.md | 46 ------- data/x-cmd/grep/SKILL.zh.md | 46 ------- data/x-cmd/grok/SKILL.md | 33 ----- data/x-cmd/grok/SKILL.zh.md | 33 ----- data/x-cmd/groovy/SKILL.md | 26 ---- data/x-cmd/groovy/SKILL.zh.md | 26 ---- data/x-cmd/gt/SKILL.md | 38 ------ data/x-cmd/gt/SKILL.zh.md | 38 ------ data/x-cmd/gtb/SKILL.md | 45 ------- data/x-cmd/gtb/SKILL.zh.md | 45 ------- data/x-cmd/hash/SKILL.md | 47 -------- data/x-cmd/hash/SKILL.zh.md | 47 -------- data/x-cmd/hashdir/SKILL.md | 39 ------ data/x-cmd/hashdir/SKILL.zh.md | 39 ------ data/x-cmd/hi/SKILL.md | 26 ---- data/x-cmd/hi/SKILL.zh.md | 26 ---- data/x-cmd/hn/SKILL.md | 47 -------- data/x-cmd/hn/SKILL.zh.md | 47 -------- data/x-cmd/host/SKILL.md | 47 -------- data/x-cmd/host/SKILL.zh.md | 47 -------- data/x-cmd/hostname/SKILL.md | 39 ------ data/x-cmd/hostname/SKILL.zh.md | 39 ------ data/x-cmd/hub/SKILL.md | 47 -------- data/x-cmd/hub/SKILL.zh.md | 47 -------- data/x-cmd/humantime/SKILL.md | 45 ------- data/x-cmd/humantime/SKILL.zh.md | 45 ------- data/x-cmd/id/SKILL.md | 45 ------- data/x-cmd/id/SKILL.zh.md | 45 ------- data/x-cmd/ip/SKILL.md | 53 -------- data/x-cmd/ip/SKILL.zh.md | 53 -------- data/x-cmd/java/SKILL.md | 28 ----- data/x-cmd/java/SKILL.zh.md | 28 ----- data/x-cmd/jina/SKILL.md | 58 --------- data/x-cmd/jina/SKILL.zh.md | 58 --------- data/x-cmd/jq/SKILL.md | 56 --------- data/x-cmd/jq/SKILL.zh.md | 56 --------- data/x-cmd/kev/SKILL.md | 36 ------ data/x-cmd/kev/SKILL.zh.md | 36 ------ data/x-cmd/kill/SKILL.md | 47 -------- data/x-cmd/kill/SKILL.zh.md | 47 -------- data/x-cmd/kimi/SKILL.md | 51 -------- data/x-cmd/kimi/SKILL.zh.md | 51 -------- data/x-cmd/last/SKILL.md | 52 -------- data/x-cmd/last/SKILL.zh.md | 52 -------- data/x-cmd/ll/SKILL.md | 37 ------ data/x-cmd/ll/SKILL.zh.md | 37 ------ data/x-cmd/llmf/SKILL.md | 51 -------- data/x-cmd/llmf/SKILL.zh.md | 51 -------- data/x-cmd/lms/SKILL.md | 38 ------ data/x-cmd/lms/SKILL.zh.md | 38 ------ data/x-cmd/ls/SKILL.md | 45 ------- data/x-cmd/ls/SKILL.zh.md | 45 ------- data/x-cmd/lsio/SKILL.md | 27 ----- data/x-cmd/lsio/SKILL.zh.md | 27 ----- data/x-cmd/lsof/SKILL.md | 47 -------- data/x-cmd/lsof/SKILL.zh.md | 47 -------- data/x-cmd/lua/SKILL.md | 46 ------- data/x-cmd/lua/SKILL.zh.md | 46 ------- data/x-cmd/luarocks/SKILL.md | 27 ----- data/x-cmd/luarocks/SKILL.zh.md | 27 ----- data/x-cmd/mac/SKILL.md | 49 -------- data/x-cmd/mac/SKILL.zh.md | 49 -------- data/x-cmd/man/SKILL.md | 46 ------- data/x-cmd/man/SKILL.zh.md | 46 ------- data/x-cmd/mankier/SKILL.md | 47 -------- data/x-cmd/mankier/SKILL.zh.md | 47 -------- data/x-cmd/mistral/SKILL.md | 39 ------ data/x-cmd/mistral/SKILL.zh.md | 39 ------ data/x-cmd/moonshot/SKILL.md | 43 ------- data/x-cmd/moonshot/SKILL.zh.md | 43 ------- data/x-cmd/mosh/SKILL.md | 40 ------ data/x-cmd/mosh/SKILL.zh.md | 40 ------ data/x-cmd/nets/SKILL.md | 40 ------ data/x-cmd/nets/SKILL.zh.md | 40 ------ data/x-cmd/node/SKILL.md | 34 ------ data/x-cmd/node/SKILL.zh.md | 34 ------ data/x-cmd/npm/SKILL.md | 27 ----- data/x-cmd/npm/SKILL.zh.md | 27 ----- data/x-cmd/npx/SKILL.md | 27 ----- data/x-cmd/npx/SKILL.zh.md | 27 ----- data/x-cmd/nu/SKILL.md | 38 ------ data/x-cmd/nu/SKILL.zh.md | 38 ------ data/x-cmd/ollama/SKILL.md | 52 -------- data/x-cmd/ollama/SKILL.zh.md | 52 -------- data/x-cmd/onsh/SKILL.md | 38 ------ data/x-cmd/onsh/SKILL.zh.md | 38 ------ data/x-cmd/openai/SKILL.md | 59 --------- data/x-cmd/openai/SKILL.zh.md | 59 --------- data/x-cmd/openclaw/SKILL.md | 58 --------- data/x-cmd/openclaw/SKILL.zh.md | 58 --------- data/x-cmd/openspec/SKILL.md | 51 -------- data/x-cmd/openspec/SKILL.zh.md | 51 -------- data/x-cmd/os/SKILL.md | 48 -------- data/x-cmd/os/SKILL.zh.md | 48 -------- data/x-cmd/osv/SKILL.md | 47 -------- data/x-cmd/osv/SKILL.zh.md | 47 -------- data/x-cmd/pandoc/SKILL.md | 47 -------- data/x-cmd/pandoc/SKILL.zh.md | 47 -------- data/x-cmd/path/SKILL.md | 46 ------- data/x-cmd/path/SKILL.zh.md | 46 ------- data/x-cmd/pb/SKILL.md | 46 ------- data/x-cmd/pb/SKILL.zh.md | 46 ------- data/x-cmd/perl/SKILL.md | 26 ---- data/x-cmd/perl/SKILL.zh.md | 26 ---- data/x-cmd/pick/SKILL.md | 46 ------- data/x-cmd/pick/SKILL.zh.md | 46 ------- data/x-cmd/ping/SKILL.md | 38 ------ data/x-cmd/ping/SKILL.zh.md | 38 ------ data/x-cmd/pip/SKILL.md | 27 ----- data/x-cmd/pip/SKILL.zh.md | 27 ----- data/x-cmd/pixi/SKILL.md | 47 -------- data/x-cmd/pixi/SKILL.zh.md | 47 -------- data/x-cmd/pkgx/SKILL.md | 46 ------- data/x-cmd/pkgx/SKILL.zh.md | 46 ------- data/x-cmd/proxy/SKILL.md | 46 ------- data/x-cmd/proxy/SKILL.zh.md | 46 ------- data/x-cmd/ps/SKILL.md | 46 ------- data/x-cmd/ps/SKILL.zh.md | 46 ------- data/x-cmd/pwsh/SKILL.md | 34 ------ data/x-cmd/pwsh/SKILL.zh.md | 34 ------ data/x-cmd/python/SKILL.md | 38 ------ data/x-cmd/python/SKILL.zh.md | 38 ------ data/x-cmd/qywx/SKILL.md | 46 ------- data/x-cmd/qywx/SKILL.zh.md | 46 ------- data/x-cmd/raku/SKILL.md | 25 ---- data/x-cmd/raku/SKILL.zh.md | 25 ---- data/x-cmd/rand/SKILL.md | 48 -------- data/x-cmd/rand/SKILL.zh.md | 48 -------- data/x-cmd/rg/SKILL.md | 51 -------- data/x-cmd/rg/SKILL.zh.md | 51 -------- data/x-cmd/route/SKILL.md | 37 ------ data/x-cmd/route/SKILL.zh.md | 37 ------ data/x-cmd/sb/SKILL.md | 38 ------ data/x-cmd/sb/SKILL.zh.md | 38 ------ data/x-cmd/scoop/SKILL.md | 48 -------- data/x-cmd/scoop/SKILL.zh.md | 48 -------- data/x-cmd/scorecard/SKILL.md | 38 ------ data/x-cmd/scorecard/SKILL.zh.md | 38 ------ data/x-cmd/sd/SKILL.md | 40 ------ data/x-cmd/sd/SKILL.zh.md | 40 ------ data/x-cmd/se/SKILL.md | 46 ------- data/x-cmd/se/SKILL.zh.md | 46 ------- data/x-cmd/sed/SKILL.md | 47 -------- data/x-cmd/sed/SKILL.zh.md | 47 -------- data/x-cmd/shodan/SKILL.md | 47 -------- data/x-cmd/shodan/SKILL.zh.md | 47 -------- data/x-cmd/sleep/SKILL.md | 52 -------- data/x-cmd/sleep/SKILL.zh.md | 52 -------- data/x-cmd/smart/SKILL.md | 47 -------- data/x-cmd/smart/SKILL.zh.md | 47 -------- data/x-cmd/speckit/SKILL.md | 41 ------- data/x-cmd/speckit/SKILL.zh.md | 41 ------- data/x-cmd/stat/SKILL.md | 46 ------- data/x-cmd/stat/SKILL.zh.md | 46 ------- data/x-cmd/str/SKILL.md | 53 -------- data/x-cmd/str/SKILL.zh.md | 53 -------- data/x-cmd/sudo/SKILL.md | 46 ------- data/x-cmd/sudo/SKILL.zh.md | 46 ------- data/x-cmd/tcsh/SKILL.md | 26 ---- data/x-cmd/tcsh/SKILL.zh.md | 26 ---- data/x-cmd/tea/SKILL.md | 30 ----- data/x-cmd/tea/SKILL.zh.md | 30 ----- data/x-cmd/tee/SKILL.md | 46 ------- data/x-cmd/tee/SKILL.zh.md | 46 ------- data/x-cmd/telegram/SKILL.md | 40 ------ data/x-cmd/telegram/SKILL.zh.md | 40 ------ data/x-cmd/termux/SKILL.md | 50 -------- data/x-cmd/termux/SKILL.zh.md | 50 -------- data/x-cmd/tldr/SKILL.md | 45 ------- data/x-cmd/tldr/SKILL.zh.md | 45 ------- data/x-cmd/tmp/SKILL.md | 46 ------- data/x-cmd/tmp/SKILL.zh.md | 46 ------- data/x-cmd/tping/SKILL.md | 38 ------ data/x-cmd/tping/SKILL.zh.md | 38 ------ data/x-cmd/uname/SKILL.md | 30 ----- data/x-cmd/uname/SKILL.zh.md | 30 ----- data/x-cmd/uninstall/SKILL.md | 45 ------- data/x-cmd/uninstall/SKILL.zh.md | 45 ------- data/x-cmd/uptime/SKILL.md | 38 ------ data/x-cmd/uptime/SKILL.zh.md | 38 ------ data/x-cmd/webtop/SKILL.md | 39 ------ data/x-cmd/webtop/SKILL.zh.md | 39 ------ data/x-cmd/whisper/SKILL.md | 51 -------- data/x-cmd/whisper/SKILL.zh.md | 51 -------- data/x-cmd/winget/SKILL.md | 38 ------ data/x-cmd/winget/SKILL.zh.md | 38 ------ data/x-cmd/wkp/SKILL.md | 51 -------- data/x-cmd/wkp/SKILL.zh.md | 51 -------- data/x-cmd/writer/SKILL.md | 51 -------- data/x-cmd/writer/SKILL.zh.md | 51 -------- data/x-cmd/ws/SKILL.md | 38 ------ data/x-cmd/ws/SKILL.zh.md | 38 ------ data/x-cmd/wttr/SKILL.md | 44 ------- data/x-cmd/wttr/SKILL.zh.md | 44 ------- data/x-cmd/yq/SKILL.md | 47 -------- data/x-cmd/yq/SKILL.zh.md | 47 -------- data/x-cmd/zhipu/SKILL.md | 39 ------ data/x-cmd/zhipu/SKILL.zh.md | 39 ------ data/x-cmd/zig/SKILL.md | 47 -------- data/x-cmd/zig/SKILL.zh.md | 47 -------- 294 files changed, 138 insertions(+), 12633 deletions(-) delete mode 100644 data/x-cmd/aider/SKILL.md delete mode 100644 data/x-cmd/aider/SKILL.zh.md delete mode 100644 data/x-cmd/arp/SKILL.md delete mode 100644 data/x-cmd/arp/SKILL.zh.md delete mode 100644 data/x-cmd/asdf/SKILL.md delete mode 100644 data/x-cmd/asdf/SKILL.zh.md delete mode 100644 data/x-cmd/assert/SKILL.md delete mode 100644 data/x-cmd/assert/SKILL.zh.md delete mode 100644 data/x-cmd/brew/SKILL.md delete mode 100644 data/x-cmd/brew/SKILL.zh.md delete mode 100644 data/x-cmd/bun/SKILL.md delete mode 100644 data/x-cmd/bun/SKILL.zh.md delete mode 100644 data/x-cmd/buse/SKILL.md delete mode 100644 data/x-cmd/buse/SKILL.zh.md delete mode 100644 data/x-cmd/bwh/SKILL.md delete mode 100644 data/x-cmd/bwh/SKILL.zh.md delete mode 100644 data/x-cmd/cal/SKILL.md delete mode 100644 data/x-cmd/cal/SKILL.zh.md delete mode 100644 data/x-cmd/cb/SKILL.md delete mode 100644 data/x-cmd/cb/SKILL.zh.md delete mode 100644 data/x-cmd/ccal/SKILL.md delete mode 100644 data/x-cmd/ccal/SKILL.zh.md delete mode 100644 data/x-cmd/ccmd/SKILL.md delete mode 100644 data/x-cmd/ccmd/SKILL.zh.md delete mode 100644 data/x-cmd/cd/SKILL.md delete mode 100644 data/x-cmd/cd/SKILL.zh.md delete mode 100644 data/x-cmd/cht/SKILL.md delete mode 100644 data/x-cmd/cht/SKILL.zh.md delete mode 100644 data/x-cmd/claude/SKILL.md delete mode 100644 data/x-cmd/claude/SKILL.zh.md delete mode 100644 data/x-cmd/codex/SKILL.md delete mode 100644 data/x-cmd/codex/SKILL.zh.md delete mode 100644 data/x-cmd/coincap/SKILL.md delete mode 100644 data/x-cmd/coincap/SKILL.zh.md delete mode 100644 data/x-cmd/cpu/SKILL.md delete mode 100644 data/x-cmd/cpu/SKILL.zh.md delete mode 100644 data/x-cmd/crush/SKILL.md delete mode 100644 data/x-cmd/crush/SKILL.zh.md delete mode 100644 data/x-cmd/ddgo/SKILL.md delete mode 100644 data/x-cmd/ddgo/SKILL.zh.md delete mode 100644 data/x-cmd/deepseek/SKILL.md delete mode 100644 data/x-cmd/deepseek/SKILL.zh.md delete mode 100644 data/x-cmd/deno/SKILL.md delete mode 100644 data/x-cmd/deno/SKILL.zh.md delete mode 100644 data/x-cmd/dingtalk/SKILL.md delete mode 100644 data/x-cmd/dingtalk/SKILL.zh.md delete mode 100644 data/x-cmd/discord/SKILL.md delete mode 100644 data/x-cmd/discord/SKILL.zh.md delete mode 100644 data/x-cmd/dns/SKILL.md delete mode 100644 data/x-cmd/dns/SKILL.zh.md delete mode 100644 data/x-cmd/docker/SKILL.md delete mode 100644 data/x-cmd/docker/SKILL.zh.md delete mode 100644 data/x-cmd/doubao/SKILL.md delete mode 100644 data/x-cmd/doubao/SKILL.zh.md delete mode 100644 data/x-cmd/elv/SKILL.md delete mode 100644 data/x-cmd/elv/SKILL.zh.md delete mode 100644 data/x-cmd/facl/SKILL.md delete mode 100644 data/x-cmd/facl/SKILL.zh.md delete mode 100644 data/x-cmd/feishu/SKILL.md delete mode 100644 data/x-cmd/feishu/SKILL.zh.md delete mode 100644 data/x-cmd/ffmpeg/SKILL.md delete mode 100644 data/x-cmd/ffmpeg/SKILL.zh.md delete mode 100644 data/x-cmd/find/SKILL.md delete mode 100644 data/x-cmd/find/SKILL.zh.md delete mode 100644 data/x-cmd/fish/SKILL.md delete mode 100644 data/x-cmd/fish/SKILL.zh.md delete mode 100644 data/x-cmd/fjo/SKILL.md delete mode 100644 data/x-cmd/fjo/SKILL.zh.md delete mode 100644 data/x-cmd/free/SKILL.md delete mode 100644 data/x-cmd/free/SKILL.zh.md delete mode 100644 data/x-cmd/fsiter/SKILL.md delete mode 100644 data/x-cmd/fsiter/SKILL.zh.md delete mode 100644 data/x-cmd/gcal/SKILL.md delete mode 100644 data/x-cmd/gcal/SKILL.zh.md delete mode 100644 data/x-cmd/gddy/SKILL.md delete mode 100644 data/x-cmd/gddy/SKILL.zh.md delete mode 100644 data/x-cmd/gemini/SKILL.md delete mode 100644 data/x-cmd/gemini/SKILL.zh.md delete mode 100644 data/x-cmd/gg/SKILL.md delete mode 100644 data/x-cmd/gg/SKILL.zh.md delete mode 100644 data/x-cmd/gh/SKILL.md delete mode 100644 data/x-cmd/gh/SKILL.zh.md delete mode 100644 data/x-cmd/gitconfig/SKILL.md delete mode 100644 data/x-cmd/gitconfig/SKILL.zh.md delete mode 100644 data/x-cmd/githook/SKILL.md delete mode 100644 data/x-cmd/githook/SKILL.zh.md delete mode 100644 data/x-cmd/gl/SKILL.md delete mode 100644 data/x-cmd/gl/SKILL.zh.md delete mode 100644 data/x-cmd/gm/SKILL.md delete mode 100644 data/x-cmd/gm/SKILL.zh.md delete mode 100644 data/x-cmd/gram/SKILL.md delete mode 100644 data/x-cmd/gram/SKILL.zh.md delete mode 100644 data/x-cmd/grep/SKILL.md delete mode 100644 data/x-cmd/grep/SKILL.zh.md delete mode 100644 data/x-cmd/grok/SKILL.md delete mode 100644 data/x-cmd/grok/SKILL.zh.md delete mode 100644 data/x-cmd/groovy/SKILL.md delete mode 100644 data/x-cmd/groovy/SKILL.zh.md delete mode 100644 data/x-cmd/gt/SKILL.md delete mode 100644 data/x-cmd/gt/SKILL.zh.md delete mode 100644 data/x-cmd/gtb/SKILL.md delete mode 100644 data/x-cmd/gtb/SKILL.zh.md delete mode 100644 data/x-cmd/hash/SKILL.md delete mode 100644 data/x-cmd/hash/SKILL.zh.md delete mode 100644 data/x-cmd/hashdir/SKILL.md delete mode 100644 data/x-cmd/hashdir/SKILL.zh.md delete mode 100644 data/x-cmd/hi/SKILL.md delete mode 100644 data/x-cmd/hi/SKILL.zh.md delete mode 100644 data/x-cmd/hn/SKILL.md delete mode 100644 data/x-cmd/hn/SKILL.zh.md delete mode 100644 data/x-cmd/host/SKILL.md delete mode 100644 data/x-cmd/host/SKILL.zh.md delete mode 100644 data/x-cmd/hostname/SKILL.md delete mode 100644 data/x-cmd/hostname/SKILL.zh.md delete mode 100644 data/x-cmd/hub/SKILL.md delete mode 100644 data/x-cmd/hub/SKILL.zh.md delete mode 100644 data/x-cmd/humantime/SKILL.md delete mode 100644 data/x-cmd/humantime/SKILL.zh.md delete mode 100644 data/x-cmd/id/SKILL.md delete mode 100644 data/x-cmd/id/SKILL.zh.md delete mode 100644 data/x-cmd/ip/SKILL.md delete mode 100644 data/x-cmd/ip/SKILL.zh.md delete mode 100644 data/x-cmd/java/SKILL.md delete mode 100644 data/x-cmd/java/SKILL.zh.md delete mode 100644 data/x-cmd/jina/SKILL.md delete mode 100644 data/x-cmd/jina/SKILL.zh.md delete mode 100644 data/x-cmd/jq/SKILL.md delete mode 100644 data/x-cmd/jq/SKILL.zh.md delete mode 100644 data/x-cmd/kev/SKILL.md delete mode 100644 data/x-cmd/kev/SKILL.zh.md delete mode 100644 data/x-cmd/kill/SKILL.md delete mode 100644 data/x-cmd/kill/SKILL.zh.md delete mode 100644 data/x-cmd/kimi/SKILL.md delete mode 100644 data/x-cmd/kimi/SKILL.zh.md delete mode 100644 data/x-cmd/last/SKILL.md delete mode 100644 data/x-cmd/last/SKILL.zh.md delete mode 100644 data/x-cmd/ll/SKILL.md delete mode 100644 data/x-cmd/ll/SKILL.zh.md delete mode 100644 data/x-cmd/llmf/SKILL.md delete mode 100644 data/x-cmd/llmf/SKILL.zh.md delete mode 100644 data/x-cmd/lms/SKILL.md delete mode 100644 data/x-cmd/lms/SKILL.zh.md delete mode 100644 data/x-cmd/ls/SKILL.md delete mode 100644 data/x-cmd/ls/SKILL.zh.md delete mode 100644 data/x-cmd/lsio/SKILL.md delete mode 100644 data/x-cmd/lsio/SKILL.zh.md delete mode 100644 data/x-cmd/lsof/SKILL.md delete mode 100644 data/x-cmd/lsof/SKILL.zh.md delete mode 100644 data/x-cmd/lua/SKILL.md delete mode 100644 data/x-cmd/lua/SKILL.zh.md delete mode 100644 data/x-cmd/luarocks/SKILL.md delete mode 100644 data/x-cmd/luarocks/SKILL.zh.md delete mode 100644 data/x-cmd/mac/SKILL.md delete mode 100644 data/x-cmd/mac/SKILL.zh.md delete mode 100644 data/x-cmd/man/SKILL.md delete mode 100644 data/x-cmd/man/SKILL.zh.md delete mode 100644 data/x-cmd/mankier/SKILL.md delete mode 100644 data/x-cmd/mankier/SKILL.zh.md delete mode 100644 data/x-cmd/mistral/SKILL.md delete mode 100644 data/x-cmd/mistral/SKILL.zh.md delete mode 100644 data/x-cmd/moonshot/SKILL.md delete mode 100644 data/x-cmd/moonshot/SKILL.zh.md delete mode 100644 data/x-cmd/mosh/SKILL.md delete mode 100644 data/x-cmd/mosh/SKILL.zh.md delete mode 100644 data/x-cmd/nets/SKILL.md delete mode 100644 data/x-cmd/nets/SKILL.zh.md delete mode 100644 data/x-cmd/node/SKILL.md delete mode 100644 data/x-cmd/node/SKILL.zh.md delete mode 100644 data/x-cmd/npm/SKILL.md delete mode 100644 data/x-cmd/npm/SKILL.zh.md delete mode 100644 data/x-cmd/npx/SKILL.md delete mode 100644 data/x-cmd/npx/SKILL.zh.md delete mode 100644 data/x-cmd/nu/SKILL.md delete mode 100644 data/x-cmd/nu/SKILL.zh.md delete mode 100644 data/x-cmd/ollama/SKILL.md delete mode 100644 data/x-cmd/ollama/SKILL.zh.md delete mode 100644 data/x-cmd/onsh/SKILL.md delete mode 100644 data/x-cmd/onsh/SKILL.zh.md delete mode 100644 data/x-cmd/openai/SKILL.md delete mode 100644 data/x-cmd/openai/SKILL.zh.md delete mode 100644 data/x-cmd/openclaw/SKILL.md delete mode 100644 data/x-cmd/openclaw/SKILL.zh.md delete mode 100644 data/x-cmd/openspec/SKILL.md delete mode 100644 data/x-cmd/openspec/SKILL.zh.md delete mode 100644 data/x-cmd/os/SKILL.md delete mode 100644 data/x-cmd/os/SKILL.zh.md delete mode 100644 data/x-cmd/osv/SKILL.md delete mode 100644 data/x-cmd/osv/SKILL.zh.md delete mode 100644 data/x-cmd/pandoc/SKILL.md delete mode 100644 data/x-cmd/pandoc/SKILL.zh.md delete mode 100644 data/x-cmd/path/SKILL.md delete mode 100644 data/x-cmd/path/SKILL.zh.md delete mode 100644 data/x-cmd/pb/SKILL.md delete mode 100644 data/x-cmd/pb/SKILL.zh.md delete mode 100644 data/x-cmd/perl/SKILL.md delete mode 100644 data/x-cmd/perl/SKILL.zh.md delete mode 100644 data/x-cmd/pick/SKILL.md delete mode 100644 data/x-cmd/pick/SKILL.zh.md delete mode 100644 data/x-cmd/ping/SKILL.md delete mode 100644 data/x-cmd/ping/SKILL.zh.md delete mode 100644 data/x-cmd/pip/SKILL.md delete mode 100644 data/x-cmd/pip/SKILL.zh.md delete mode 100644 data/x-cmd/pixi/SKILL.md delete mode 100644 data/x-cmd/pixi/SKILL.zh.md delete mode 100644 data/x-cmd/pkgx/SKILL.md delete mode 100644 data/x-cmd/pkgx/SKILL.zh.md delete mode 100644 data/x-cmd/proxy/SKILL.md delete mode 100644 data/x-cmd/proxy/SKILL.zh.md delete mode 100644 data/x-cmd/ps/SKILL.md delete mode 100644 data/x-cmd/ps/SKILL.zh.md delete mode 100644 data/x-cmd/pwsh/SKILL.md delete mode 100644 data/x-cmd/pwsh/SKILL.zh.md delete mode 100644 data/x-cmd/python/SKILL.md delete mode 100644 data/x-cmd/python/SKILL.zh.md delete mode 100644 data/x-cmd/qywx/SKILL.md delete mode 100644 data/x-cmd/qywx/SKILL.zh.md delete mode 100644 data/x-cmd/raku/SKILL.md delete mode 100644 data/x-cmd/raku/SKILL.zh.md delete mode 100644 data/x-cmd/rand/SKILL.md delete mode 100644 data/x-cmd/rand/SKILL.zh.md delete mode 100644 data/x-cmd/rg/SKILL.md delete mode 100644 data/x-cmd/rg/SKILL.zh.md delete mode 100644 data/x-cmd/route/SKILL.md delete mode 100644 data/x-cmd/route/SKILL.zh.md delete mode 100644 data/x-cmd/sb/SKILL.md delete mode 100644 data/x-cmd/sb/SKILL.zh.md delete mode 100644 data/x-cmd/scoop/SKILL.md delete mode 100644 data/x-cmd/scoop/SKILL.zh.md delete mode 100644 data/x-cmd/scorecard/SKILL.md delete mode 100644 data/x-cmd/scorecard/SKILL.zh.md delete mode 100644 data/x-cmd/sd/SKILL.md delete mode 100644 data/x-cmd/sd/SKILL.zh.md delete mode 100644 data/x-cmd/se/SKILL.md delete mode 100644 data/x-cmd/se/SKILL.zh.md delete mode 100644 data/x-cmd/sed/SKILL.md delete mode 100644 data/x-cmd/sed/SKILL.zh.md delete mode 100644 data/x-cmd/shodan/SKILL.md delete mode 100644 data/x-cmd/shodan/SKILL.zh.md delete mode 100644 data/x-cmd/sleep/SKILL.md delete mode 100644 data/x-cmd/sleep/SKILL.zh.md delete mode 100644 data/x-cmd/smart/SKILL.md delete mode 100644 data/x-cmd/smart/SKILL.zh.md delete mode 100644 data/x-cmd/speckit/SKILL.md delete mode 100644 data/x-cmd/speckit/SKILL.zh.md delete mode 100644 data/x-cmd/stat/SKILL.md delete mode 100644 data/x-cmd/stat/SKILL.zh.md delete mode 100644 data/x-cmd/str/SKILL.md delete mode 100644 data/x-cmd/str/SKILL.zh.md delete mode 100644 data/x-cmd/sudo/SKILL.md delete mode 100644 data/x-cmd/sudo/SKILL.zh.md delete mode 100644 data/x-cmd/tcsh/SKILL.md delete mode 100644 data/x-cmd/tcsh/SKILL.zh.md delete mode 100644 data/x-cmd/tea/SKILL.md delete mode 100644 data/x-cmd/tea/SKILL.zh.md delete mode 100644 data/x-cmd/tee/SKILL.md delete mode 100644 data/x-cmd/tee/SKILL.zh.md delete mode 100644 data/x-cmd/telegram/SKILL.md delete mode 100644 data/x-cmd/telegram/SKILL.zh.md delete mode 100644 data/x-cmd/termux/SKILL.md delete mode 100644 data/x-cmd/termux/SKILL.zh.md delete mode 100644 data/x-cmd/tldr/SKILL.md delete mode 100644 data/x-cmd/tldr/SKILL.zh.md delete mode 100644 data/x-cmd/tmp/SKILL.md delete mode 100644 data/x-cmd/tmp/SKILL.zh.md delete mode 100644 data/x-cmd/tping/SKILL.md delete mode 100644 data/x-cmd/tping/SKILL.zh.md delete mode 100644 data/x-cmd/uname/SKILL.md delete mode 100644 data/x-cmd/uname/SKILL.zh.md delete mode 100644 data/x-cmd/uninstall/SKILL.md delete mode 100644 data/x-cmd/uninstall/SKILL.zh.md delete mode 100644 data/x-cmd/uptime/SKILL.md delete mode 100644 data/x-cmd/uptime/SKILL.zh.md delete mode 100644 data/x-cmd/webtop/SKILL.md delete mode 100644 data/x-cmd/webtop/SKILL.zh.md delete mode 100644 data/x-cmd/whisper/SKILL.md delete mode 100644 data/x-cmd/whisper/SKILL.zh.md delete mode 100644 data/x-cmd/winget/SKILL.md delete mode 100644 data/x-cmd/winget/SKILL.zh.md delete mode 100644 data/x-cmd/wkp/SKILL.md delete mode 100644 data/x-cmd/wkp/SKILL.zh.md delete mode 100644 data/x-cmd/writer/SKILL.md delete mode 100644 data/x-cmd/writer/SKILL.zh.md delete mode 100644 data/x-cmd/ws/SKILL.md delete mode 100644 data/x-cmd/ws/SKILL.zh.md delete mode 100644 data/x-cmd/wttr/SKILL.md delete mode 100644 data/x-cmd/wttr/SKILL.zh.md delete mode 100644 data/x-cmd/yq/SKILL.md delete mode 100644 data/x-cmd/yq/SKILL.zh.md delete mode 100644 data/x-cmd/zhipu/SKILL.md delete mode 100644 data/x-cmd/zhipu/SKILL.zh.md delete mode 100644 data/x-cmd/zig/SKILL.md delete mode 100644 data/x-cmd/zig/SKILL.zh.md diff --git a/data/minimax/minimax-multimodal-toolkit/SKILL.md b/data/minimax/minimax-multimodal-toolkit/SKILL.md index 765a730..b7e856f 100644 --- a/data/minimax/minimax-multimodal-toolkit/SKILL.md +++ b/data/minimax/minimax-multimodal-toolkit/SKILL.md @@ -76,6 +76,37 @@ Before running any script, check if `MINIMAX_API_KEY` is set in the environment. 1. Ask the user to provide their MiniMax API key 2. Instruct and help user to set it via `export MINIMAX_API_KEY="sk-..."` in their terminal or add it to their shell profile (`~/.zshrc` / `~/.bashrc`) for persistence +## Plan Limits & Quotas + +**IMPORTANT — Always respect the user's plan limits before generating content.** If the user's quota is exhausted or insufficient, warn them before proceeding. + +### Standard Plans + +| Capability | Starter | Plus | Max | +|---|---|---|---| +| M2.7 (chat) | 600 req/5h | 1,500 req/5h | 4,500 req/5h | +| Speech 2.8 | — | 4,000 chars/day | 11,000 chars/day | +| image-01 | — | 50 images/day | 120 images/day | +| Hailuo-2.3-Fast 768P 6s | — | — | 2 videos/day | +| Hailuo-2.3 768P 6s | — | — | 2 videos/day | +| Music-2.5 | — | — | 4 songs/day (≤5 min each) | + +### High-Speed Plans + +| Capability | Plus-HS | Max-HS | Ultra-HS | +|---|---|---|---| +| M2.7-highspeed (chat) | 1,500 req/5h | 4,500 req/5h | 30,000 req/5h | +| Speech 2.8 | 9,000 chars/day | 19,000 chars/day | 50,000 chars/day | +| image-01 | 100 images/day | 200 images/day | 800 images/day | +| Hailuo-2.3-Fast 768P 6s | — | 3 videos/day | 5 videos/day | +| Hailuo-2.3 768P 6s | — | 3 videos/day | 5 videos/day | +| Music-2.5 | — | 7 songs/day (≤5 min each) | 15 songs/day (≤5 min each) | + +**Key quota constraints:** +- **Video resolution: 768P only** — 1080P is not available on any plan +- **Video duration: 6s** — all plan quotas are counted in 6-second units +- **Video quota is very limited** (2–5/day depending on plan) — always confirm with the user before generating video + ## Key Capabilities | Capability | Description | Entry point | @@ -384,40 +415,30 @@ bash scripts/image/generate_image.sh \ | User intent | Script to use | |-------------|---------------| -| Default / no special request | `scripts/video/generate_video.sh` (single segment, **10s, 768P**) | +| Default / no special request | `scripts/video/generate_video.sh` (single segment, **6s, 768P**) | | User explicitly asks for "long video", "multi-scene", "story", or duration > 10s | `scripts/video/generate_long_video.sh` (multi-segment) | -**Default behavior:** Always use single-segment `generate_video.sh` with **duration 10s and resolution 768P** unless the user explicitly asks for a long video, multi-scene video, or specifies a total duration exceeding 10 seconds. Do NOT automatically split into multiple segments — a single 10s video is the standard output. Only use `generate_long_video.sh` when the user clearly needs multi-scene or longer content. +**Default behavior:** Always use single-segment `generate_video.sh` with **duration 6s and resolution 768P** unless the user explicitly asks for a long video or multi-scene video. Do NOT automatically split into multiple segments — a single 6s video is the standard output. Only use `generate_long_video.sh` when the user clearly needs multi-scene or longer content. Entry point (single video): `scripts/video/generate_video.sh` Entry point (long/multi-scene): `scripts/video/generate_long_video.sh` ### Video Model Constraints (MUST follow) -**Duration limits by model and resolution:** - -| Model | 720P | 768P | 1080P | -|-------|------|------|-------| -| MiniMax-Hailuo-2.3 | - | 6s or **10s** | 6s only | -| MiniMax-Hailuo-2.3-Fast | - | 6s or **10s** | 6s only | -| MiniMax-Hailuo-02 | - | 6s or **10s** | 6s only | -| T2V-01 / T2V-01-Director | 6s only | - | - | -| I2V-01 / I2V-01-Director / I2V-01-live | 6s only | - | - | -| S2V-01 (ref) | 6s only | - | - | - -**Resolution options by model and duration:** +**Supported resolutions and durations by model:** -| Model | 6s | 10s | -|-------|-----|-----| -| MiniMax-Hailuo-2.3 | 768P (default), 1080P | 768P only | -| MiniMax-Hailuo-2.3-Fast | 768P (default), 1080P | 768P only | -| MiniMax-Hailuo-02 | 512P, 768P (default), 1080P | 512P, 768P (default) | -| Other models | 720P (default) | Not supported | +| Model | Resolution | Duration | +|-------|-----------|----------| +| MiniMax-Hailuo-2.3 | 768P only | 6s or 10s | +| MiniMax-Hailuo-2.3-Fast | 768P only | 6s or 10s | +| MiniMax-Hailuo-02 | 512P, 768P (default) | 6s or 10s | +| T2V-01 / T2V-01-Director | 720P | 6s only | +| I2V-01 / I2V-01-Director / I2V-01-live | 720P | 6s only | +| S2V-01 (ref) | 720P | 6s only | **Key rules:** -- **Default: 10s + 768P** (best balance of length and quality for MiniMax-Hailuo-2.3) -- 1080P only supports 6s duration — if user requests 1080P, set `--duration 6` -- 10s duration only works with 768P (or 512P on Hailuo-02) — never combine 10s + 1080P +- **Default: 6s + 768P** — plan quotas are counted in 6-second units; use 6s unless user explicitly requests 10s +- **1080P is NOT supported** on any plan — always use 768P for Hailuo-2.3/2.3-Fast - Older models (T2V-01, I2V-01, S2V-01) only support 6s at 720P ### IMPORTANT: Prompt Optimization (MUST follow before generating any video) @@ -443,19 +464,12 @@ Before calling any video generation script, you MUST optimize the user's prompt 6. **For multi-segment long videos**: Each segment's prompt must be self-contained and optimized individually. The i2v segments (segment 2+) should describe motion/change relative to the previous segment's ending frame. ```bash -# Text-to-video (default: 10s, 768P) +# Text-to-video (default: 6s, 768P) bash scripts/video/generate_video.sh \ --mode t2v \ --prompt "A golden retriever puppy bounds toward the camera on a sunlit grass path, [跟随] tracking shot, warm golden hour, shallow depth of field, joyful" \ --output minimax-output/puppy.mp4 -# Text-to-video with 1080P (must use --duration 6) -bash scripts/video/generate_video.sh \ - --mode t2v \ - --prompt "A golden retriever puppy bounds toward the camera" \ - --duration 6 --resolution 1080P \ - --output minimax-output/puppy_hd.mp4 - # Image-to-video (prompt focuses on MOTION, not image content) bash scripts/video/generate_video.sh \ --mode i2v \ @@ -480,7 +494,7 @@ bash scripts/video/generate_video.sh \ ### Long-form Video (Multi-scene) -Multi-scene long videos chain segments together: the first segment generates via text-to-video (t2v), then each subsequent segment uses the last frame of the previous segment as its first frame (i2v). Segments are joined with crossfade transitions for smooth continuity. Default is 10 seconds per segment. +Multi-scene long videos chain segments together: the first segment generates via text-to-video (t2v), then each subsequent segment uses the last frame of the previous segment as its first frame (i2v). Segments are joined with crossfade transitions for smooth continuity. Default is 6 seconds per segment. **Workflow:** 1. Segment 1: t2v — generated purely from the optimized text prompt @@ -493,10 +507,10 @@ Multi-scene long videos chain segments together: the first segment generates via - Segment 1 (t2v): Full scene description with subject, scene, camera, atmosphere - Segment 2+ (i2v): Focus on **what changes and moves** from the previous ending frame. Do NOT repeat the visual description — the first frame already provides it - Maintain visual consistency: keep lighting, color grading, and style keywords consistent across segments -- Each segment covers only 10 seconds of action — keep it focused +- Each segment covers only 6 seconds of action — keep it focused ```bash -# Example: 3-segment story with optimized per-segment prompts (default: 10s/segment, 768P) +# Example: 3-segment story with optimized per-segment prompts (default: 6s/segment, 768P) bash scripts/video/generate_long_video.sh \ --scenes \ "A lone astronaut stands on a red desert planet surface, wind blowing dust particles, [推进] slow push in toward the visor, dramatic rim lighting, cinematic sci-fi atmosphere" \ @@ -508,7 +522,7 @@ bash scripts/video/generate_long_video.sh \ # With custom settings bash scripts/video/generate_long_video.sh \ --scenes "Scene 1 prompt" "Scene 2 prompt" \ - --segment-duration 10 \ + --segment-duration 6 \ --resolution 768P \ --crossfade 0.5 \ --music-prompt "calm ambient background music" \ @@ -539,8 +553,8 @@ bash scripts/video/generate_template_video.sh \ | Mode | Default Model | Default Duration | Default Resolution | Notes | |------|--------------|-----------------|-------------------|-------| -| t2v | MiniMax-Hailuo-2.3 | 10s | 768P | Latest text-to-video | -| i2v | MiniMax-Hailuo-2.3 | 10s | 768P | Latest image-to-video | +| t2v | MiniMax-Hailuo-2.3 | 6s | 768P | Latest text-to-video | +| i2v | MiniMax-Hailuo-2.3 | 6s | 768P | Latest image-to-video | | sef | MiniMax-Hailuo-02 | 6s | 768P | Start-end frame | | ref | S2V-01 | 6s | 720P | Subject reference, 6s only | diff --git a/data/minimax/minimax-multimodal-toolkit/scripts/image/generate_image.sh b/data/minimax/minimax-multimodal-toolkit/scripts/image/generate_image.sh index 369a514..04782b9 100755 --- a/data/minimax/minimax-multimodal-toolkit/scripts/image/generate_image.sh +++ b/data/minimax/minimax-multimodal-toolkit/scripts/image/generate_image.sh @@ -44,7 +44,7 @@ image_to_data_url() { local mime mime="$(file -b --mime-type "$path" 2>/dev/null)" || mime="image/jpeg" local b64 - b64="$(base64 < "$path")" + b64="$(base64 -w 0 < "$path")" echo "data:${mime};base64,${b64}" } @@ -57,6 +57,78 @@ resolve_image() { esac } +# ============================================================================ +# Payload builder — avoids command-line length limits on Windows +# Uses temp files for jq when the payload may contain large base64 data. +# ============================================================================ + +# Build JSON payload, writing large fields (base64 image data) to temp files +# to avoid Windows cmd.exe argument-length limits (~32KB). +build_payload() { + local model="$1" prompt="$2" response_format="$3" n="$4" + local prompt_optimizer="$5" aigc_watermark="$6" + local aspect_ratio="$7" width="$8" height="$9" seed="${10:-}" + local ref_image="${11:-}" + + # Start with base payload using temp file to avoid long command lines + local base_tmp + base_tmp="$(mktemp)" + trap "rm -f '$base_tmp'" EXIT INT TERM HUP + + jq -n \ + --arg model "$model" \ + --arg prompt "$prompt" \ + --arg rf "$response_format" \ + --argjson n "$n" \ + --argjson po "$prompt_optimizer" \ + --argjson aw "$aigc_watermark" \ + '{model: $model, prompt: $prompt, response_format: $rf, n: $n, prompt_optimizer: $po, aigc_watermark: $aw}' \ + > "$base_tmp" + + # Add optional fields, each via temp file to stay within Windows arg limits + if [[ -n "$aspect_ratio" ]]; then + local tmp2; tmp2="$(mktemp)"; trap "rm -f '$base_tmp' '$tmp2'" EXIT INT TERM HUP + jq --arg ar "$aspect_ratio" '. + {aspect_ratio: $ar}' "$base_tmp" > "$tmp2" + mv "$tmp2" "$base_tmp" + fi + if [[ -n "$width" ]]; then + local tmp2; tmp2="$(mktemp)"; trap "rm -f '$base_tmp' '$tmp2'" EXIT INT TERM HUP + jq --argjson w "$width" '. + {width: $w}' "$base_tmp" > "$tmp2" + mv "$tmp2" "$base_tmp" + fi + if [[ -n "$height" ]]; then + local tmp2; tmp2="$(mktemp)"; trap "rm -f '$base_tmp' '$tmp2'" EXIT INT TERM HUP + jq --argjson h "$height" '. + {height: $h}' "$base_tmp" > "$tmp2" + mv "$tmp2" "$base_tmp" + fi + if [[ -n "$seed" ]]; then + local tmp2; tmp2="$(mktemp)"; trap "rm -f '$base_tmp' '$tmp2'" EXIT INT TERM HUP + jq --argjson s "$seed" '. + {seed: $s}' "$base_tmp" > "$tmp2" + mv "$tmp2" "$base_tmp" + fi + + # Subject reference (i2i mode) — build via temp file to avoid huge command-line args + if [[ -n "$ref_image" ]]; then + local img_url + img_url="$(resolve_image "$ref_image")" + # Create temp files and set traps separately to avoid set -u issues + local ref_tmp; ref_tmp="$(mktemp)" + trap "rm -f '$base_tmp' '$ref_tmp'" EXIT INT TERM HUP + local url_tmp; url_tmp="$(mktemp)"; trap "rm -f '$base_tmp' '$ref_tmp' '$url_tmp'" EXIT INT TERM HUP + # Write URL to temp file to avoid long-argument issues, then build JSON + echo -n "$img_url" > "$url_tmp" + # Use jq -s to collect all lines (handles base64 with embedded newlines), take first element + jq -Rs 'split("\n")[0] | {type: "character", image_file: .}' "$url_tmp" > "$ref_tmp" + local tmp2; tmp2="$(mktemp)"; trap "rm -f '$base_tmp' '$ref_tmp' '$url_tmp' '$tmp2'" EXIT INT TERM HUP + jq --slurpfile ref "$ref_tmp" '. + {subject_reference: $ref}' "$base_tmp" > "$tmp2" + mv "$tmp2" "$base_tmp" + fi + + cat "$base_tmp" + rm -f "$base_tmp" + trap - EXIT INT TERM HUP +} + # ============================================================================ # Main # ============================================================================ @@ -107,7 +179,7 @@ Options: -n, --count N Number of images to generate (1-9, default: 1) --seed N Random seed for reproducibility --prompt-optimizer Enable automatic prompt optimization - --aigc-watermark Add AIGC watermark to generated images + --aigc-watermark Add AIGC watermark to generated images --ref-image FILE Character reference image (local file or URL, i2i mode) --response-format FMT Response format: url (default), base64 --no-download Don't download, just print URL(s) @@ -144,31 +216,13 @@ USAGE echo "Error: -n must be between 1 and 9" >&2; exit 1 fi - # Build payload + # Build payload using temp-file method (avoids Windows cmd.exe arg-length limit) local payload - payload=$(jq -n \ - --arg model "$model" \ - --arg prompt "$prompt" \ - --arg rf "$response_format" \ - --argjson n "$n" \ - --argjson po "$prompt_optimizer" \ - --argjson aw "$aigc_watermark" \ - '{model: $model, prompt: $prompt, response_format: $rf, n: $n, prompt_optimizer: $po, aigc_watermark: $aw}') - - [[ -n "$aspect_ratio" ]] && payload=$(echo "$payload" | jq --arg ar "$aspect_ratio" '. + {aspect_ratio: $ar}') - [[ -n "$width" ]] && payload=$(echo "$payload" | jq --argjson w "$width" '. + {width: $w}') - [[ -n "$height" ]] && payload=$(echo "$payload" | jq --argjson h "$height" '. + {height: $h}') - [[ -n "$seed" ]] && payload=$(echo "$payload" | jq --argjson s "$seed" '. + {seed: $s}') - - # Subject reference (i2i mode) - if [[ "$mode" == "i2i" ]]; then - if [[ -z "$ref_image" ]]; then - echo "Error: --ref-image is required for i2i mode" >&2; exit 1 - fi - local img_url - img_url="$(resolve_image "$ref_image")" - payload=$(echo "$payload" | jq --arg img "$img_url" '. + {subject_reference: [{type: "character", image_file: $img}]}') - fi + payload=$(build_payload \ + "$model" "$prompt" "$response_format" "$n" \ + "$prompt_optimizer" "$aigc_watermark" \ + "$aspect_ratio" "$width" "$height" "$seed" \ + "$ref_image") local api_host="${MINIMAX_API_HOST:-https://api.minimaxi.com}" local api_url="${api_host}/v1/image_generation" @@ -177,13 +231,18 @@ USAGE echo "Model: $model" echo "Generating $n image(s)..." + # Write payload to temp file to avoid command-line length limits + local payload_tmp; payload_tmp="$(mktemp)" + trap "rm -f '$payload_tmp'" EXIT INT TERM HUP + echo -n "$payload" > "$payload_tmp" + local raw_output http_code response raw_output="$(curl -s -w "\n%{http_code}" \ -X POST "$api_url" \ -H "Authorization: Bearer ${MINIMAX_API_KEY}" \ -H "Content-Type: application/json" \ --max-time 120 \ - -d "$payload" 2>/dev/null)" || { + -d "@$payload_tmp" 2>/dev/null)" || { echo "Error: curl request failed" >&2 exit 1 } @@ -203,6 +262,7 @@ USAGE local status_msg status_msg="$(echo "$response" | jq -r '.base_resp.status_msg // "Unknown error"')" echo "Error: API error (code $status_code): $status_msg" >&2 + echo "Full response: $response" >&2 exit 1 fi diff --git a/data/x-cmd/aider/SKILL.md b/data/x-cmd/aider/SKILL.md deleted file mode 100644 index 7b680f6..0000000 --- a/data/x-cmd/aider/SKILL.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: aider -description: > - AI pair-programming tool that allows you to code with LLMs directly in your terminal and Git repository. - Core Scenario: When the user wants to collaborate with an AI on code modifications, automated commits, or whole-repo context tasks. -license: MIT ---- - -# aider - AI Pair-Programming in Terminal - -`aider` is an AI pair-programming tool that enables developers to collaborate with high-end LLMs (Claude 3.7, DeepSeek R1, GPT-4o, etc.) directly in the terminal. It understands entire Git repositories and can perform multi-file changes and automated commits. - -## When to Activate -- When the user wants to start an AI-assisted pair-programming session. -- When performing complex code refactoring or modifications across multiple files. -- When the user needs automated, semantic Git commit messages for AI-generated changes. -- When utilizing "Voice-to-Code" or embedding external content (screenshots/webpages) into the coding context. - -## Core Principles & Rules -- **Git Integration**: Aider is most effective when run inside a Git repository. -- **Model Versatility**: Supports a wide range of models (Claude, DeepSeek, OpenAI, local LLMs). -- **Automation**: Can automatically run lints and tests to verify changes. - -## Additional Scenarios -- **Whole-Repo Context**: Automatically generates a map of the repository to help the LLM understand complex architectures. -- **Multi-File Editing**: Seamlessly handles modifications that span multiple source files. - -## Patterns & Examples - -### Launch Aider -```bash -# Start an interactive pair-programming session -x aider -``` - -## Checklist -- [ ] Verify that the user is inside a Git repository for optimal performance. -- [ ] Confirm the target LLM is configured and accessible. -- [ ] Check if additional dependencies (like for voice input) are needed. diff --git a/data/x-cmd/aider/SKILL.zh.md b/data/x-cmd/aider/SKILL.zh.md deleted file mode 100644 index f7829fe..0000000 --- a/data/x-cmd/aider/SKILL.zh.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: aider -description: > - AI 结对编程工具,允许你在终端和 Git 仓库中直接与大模型合作编写代码。 - 核心场景:当用户想要与 AI 协作进行代码修改、自动提交或处理涉及全仓库上下文的任务时。 -license: MIT ---- - -# aider - 终端中的 AI 结对编程 - -`aider` 是一款 AI 结对编程工具,使开发者能够直接在终端中与高端大模型(Claude 3.7, DeepSeek R1, GPT-4o 等)协作。它能理解整个 Git 仓库,并执行跨文件修改和自动提交。 - -## 激活时机 -- 当用户想要启动 AI 辅助的结对编程会话时。 -- 当执行复杂的代码重构或跨多个文件的修改时。 -- 当用户需要为 AI 生成的更改提供自动化的、语义化的 Git 提交信息时。 -- 当利用“语音转代码”或将外部内容(截图/网页)嵌入到编码上下文中时。 - -## 核心原则与规范 -- **Git 集成**: Aider 在 Git 仓库中运行时效果最佳。 -- **模型多样性**: 支持广泛的模型(Claude, DeepSeek, OpenAI, 本地模型)。 -- **自动化**: 可以自动运行 lint 和测试以验证更改。 - -## 补充场景 -- **全仓库上下文**: 自动生成仓库地图,帮助大模型理解复杂的架构。 -- **多文件编辑**: 无缝处理跨多个源文件的修改。 - -## 实战示例 - -### 启动 Aider -```bash -# 启动交互式结对编程会话 -x aider -``` - -## 交付验证清单 -- [ ] 验证用户是否在 Git 仓库中以获得最佳性能。 -- [ ] 确认目标大模型已配置且可访问。 -- [ ] 检查是否需要额外的依赖项(如用于语音输入)。 diff --git a/data/x-cmd/arp/SKILL.md b/data/x-cmd/arp/SKILL.md deleted file mode 100644 index ae7a1a4..0000000 --- a/data/x-cmd/arp/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: arp -description: > - Display the system's ARP cache table with support for CSV, TSV, and interactive TUI formats. - Core Scenario: When the user needs to audit IP-to-MAC address mappings on the local network. -license: MIT ---- - -# arp - ARP Cache Table Viewer - -The `arp` module provides an enhanced way to view the system's ARP cache, which maps IP addresses to physical MAC addresses. It supports structured data exports and an interactive TUI for exploration. - -## When to Activate -- When identifying devices on the local network via their MAC addresses. -- When auditing the network for suspicious or incomplete ARP entries. -- When exporting the ARP table to JSON, CSV, or TSV for automated analysis. - -## Core Principles & Rules -- **Interactive TUI**: Automatically launches a TUI app (`--app`) if connected to a terminal. -- **Vendor Discovery**: Shows hardware vendor/organization information for MAC addresses by default. -- **Data Processing**: Use `--csv` or `--tsv` for easy integration with other tools. - -## Patterns & Examples - -### Interactive Browsing -```bash -# Open the ARP table in an interactive TUI -x arp --app -``` - -### Export to CSV -```bash -# Get the full ARP cache as a CSV -x arp --csv -``` - -### Show All Entries -```bash -# Include incomplete entries in the ARP table view -x arp --all -``` - -## Checklist -- [ ] Confirm if the user needs vendor information included. -- [ ] Verify if an interactive view or static data export is required. diff --git a/data/x-cmd/arp/SKILL.zh.md b/data/x-cmd/arp/SKILL.zh.md deleted file mode 100644 index 41130b2..0000000 --- a/data/x-cmd/arp/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: arp -description: > - 显示系统的 ARP 缓存表,支持 CSV, TSV 和交互式 TUI 格式。 - 核心场景:当用户需要审核本地网络上的 IP 到 MAC 地址映射时。 -license: MIT ---- - -# arp - ARP 缓存表查看器 - -`arp` 模块提供了一种增强的方式来查看系统的 ARP 缓存(将 IP 地址映射到物理 MAC 地址)。它支持结构化数据导出和用于探索的交互式 TUI。 - -## 激活时机 -- 当通过 MAC 地址识别本地网络上的设备时。 -- 当审核网络中是否存在可疑或不完整的 ARP 条目时。 -- 当为自动化分析导出 ARP 表到 JSON, CSV 或 TSV 时。 - -## 核心原则与规范 -- **交互式 TUI**: 连接到终端时自动启动 TUI 应用 (`--app`)。 -- **厂商识别**: 默认显示 MAC 地址对应的硬件厂商/组织信息。 -- **数据处理**: 使用 `--csv` 或 `--tsv` 以便于与其他工具集成。 - -## 实战示例 - -### 交互式浏览 -```bash -# 在交互式 TUI 中打开 ARP 表 -x arp --app -``` - -### 导出为 CSV -```bash -# 以 CSV 格式获取完整的 ARP 缓存 -x arp --csv -``` - -### 显示所有条目 -```bash -# 在 ARP 表视图中包含不完整的条目 -x arp --all -``` - -## 交付验证清单 -- [ ] 确认用户是否需要包含厂商信息。 -- [ ] 验证需要的是交互式视图还是静态数据导出。 diff --git a/data/x-cmd/asdf/SKILL.md b/data/x-cmd/asdf/SKILL.md deleted file mode 100644 index c161f2f..0000000 --- a/data/x-cmd/asdf/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: asdf -description: > - Extendable version manager for runtimes (Node.js, Ruby, etc.) with enhanced x-cmd shortcuts. - Core Scenario: When the user needs to manage, install, or switch between multiple versions of development environments. -license: MIT ---- - -# asdf - Multiversion Runtime Manager - -The `asdf` module enhances the standard asdf version manager by providing simplified x-cmd style commands for installing and switching environment runtimes. - -## When to Activate -- When the user wants to install a new programming language environment (e.g., Node.js, Elixir). -- When switching between different versions of a tool globally or per project. -- When managing asdf plugins or updating the asdf core. - -## Core Principles & Rules -- **Enhanced Flow**: Use `use` to automatically add a plugin, install the latest version, and set it as global. -- **Environment State**: Use `--activate` and `--deactivate` to control if asdf managed binaries are in the current shell's PATH. -- **Zero-Setup**: Automatically handles asdf installation if it's missing. - -## Patterns & Examples - -### Install and Use -```bash -# Install the latest Node.js and set it as global -x asdf use nodejs -``` - -### Managed Uninstallation -```bash -# Uninstall a runtime and remove its corresponding asdf plugin -x asdf unuse nodejs -``` - -### Search Plugins -```bash -# Interactively search for available software/plugins to install -x asdf -``` - -## Checklist -- [ ] Confirm the target runtime (e.g., nodejs, python). -- [ ] Verify if the version should be global or specific to a directory. -- [ ] Ensure Git is available as it's required for most asdf plugins. diff --git a/data/x-cmd/asdf/SKILL.zh.md b/data/x-cmd/asdf/SKILL.zh.md deleted file mode 100644 index ea7da0c..0000000 --- a/data/x-cmd/asdf/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: asdf -description: > - 运行环境(Node.js, Ruby 等)的可扩展版本管理器,具有增强的 x-cmd 快捷方式。 - 核心场景:当用户需要管理、安装或在多个版本的开发环境之间切换时。 -license: MIT ---- - -# asdf - 多版本运行环境管理器 - -`asdf` 模块通过提供简化的 x-cmd 风格命令来安装和切换环境运行时,增强了标准的 asdf 版本管理器。 - -## 激活时机 -- 当用户想要安装新的编程语言环境(如 Node.js, Elixir)时。 -- 当全局或按项目在工具的不同版本之间进行切换时。 -- 当管理 asdf 插件或更新 asdf 核心时。 - -## 核心原则与规范 -- **增强流**: 使用 `use` 自动添加插件、安装最新版本并将其设置为全局版本。 -- **环境状态**: 使用 `--activate` 和 `--deactivate` 控制 asdf 管理的二进制文件是否处于当前 Shell 的 PATH 中。 -- **零设置**: 如果 asdf 缺失,会自动处理其安装。 - -## 实战示例 - -### 安装并使用 -```bash -# 安装最新的 Node.js 并将其设置为全局版本 -x asdf use nodejs -``` - -### 受控卸载 -```bash -# 卸载运行时并移除其对应的 asdf 插件 -x asdf unuse nodejs -``` - -### 搜索插件 -```bash -# 交互式地搜索可安装的软件/插件 -x asdf -``` - -## 交付验证清单 -- [ ] 确认目标运行时(例如 nodejs, python)。 -- [ ] 验证版本应该是全局的还是特定于目录的。 -- [ ] 确保 Git 可用,因为大多数 asdf 插件都需要它。 diff --git a/data/x-cmd/assert/SKILL.md b/data/x-cmd/assert/SKILL.md deleted file mode 100644 index aa0a97d..0000000 --- a/data/x-cmd/assert/SKILL.md +++ /dev/null @@ -1,55 +0,0 @@ ---- -name: assert -description: > - Test assertion module for validating expected behaviors, outputs, and environment states. - Core Scenario: When the user is writing tests or scripts that require verification of commands, files, or variables. -license: MIT ---- - -# assert - Robust Test Assertions - -The `assert` module is a powerful tool for verifying that test results meet expectations. It is widely used in automated testing and CI/CD pipelines to ensure code correctness. - -## When to Activate -- When writing shell scripts that need to verify command success or failure. -- When performing regression tests on file systems (checks for non-empty, existence, etc.). -- When validating variable states or detecting global variable leaks. -- When checking stdout/stderr against expected patterns or content. - -## Core Principles & Rules -- **Exit Code Focus**: Commands return 0 on success and non-zero on failure to break script execution if combined with `set -e`. -- **Negation Support**: Use `!` for negative command checks or `^` for negative type/file checks. -- **Bulk Verification**: Many subcommands support checking multiple values or files in one go. - -## Additional Scenarios -- **Variable Snapshots**: Use `var save` and `var cmp` to detect unintended global variable modifications in functions. -- **Output Matching**: Use `stdout` with heredocs to verify complex multi-line command outputs. - -## Patterns & Examples - -### Command and Type Verification -```bash -# Verify command success and multiple integer values -x assert true [ 1 -eq 1 ] -x assert is-int 10 20 30 -``` - -### File System Checks -```bash -# Ensure specific files exist and are not empty -x assert is-file /etc/hosts -x assert is-nonempty ./log.txt -``` - -### Detecting Variable Leaks -```bash -# Capture state, run function, and check for leaks -x assert var save -my_function -x assert var cmp -``` - -## Checklist -- [ ] Ensure the correct operator (`is-file`, `is-int`, etc.) is used. -- [ ] Confirm if negation (`!` or `^`) is required for the test logic. -- [ ] Verify that multi-line expectations match the stdin/heredoc format. diff --git a/data/x-cmd/assert/SKILL.zh.md b/data/x-cmd/assert/SKILL.zh.md deleted file mode 100644 index 844d742..0000000 --- a/data/x-cmd/assert/SKILL.zh.md +++ /dev/null @@ -1,55 +0,0 @@ ---- -name: assert -description: > - 用于验证预期行为、输出和环境状态的测试断言模块。 - 核心场景:当用户正在编写需要验证命令、文件或变量的测试或脚本时。 -license: MIT ---- - -# assert - 健壮的测试断言 - -`assert` 模块是一个功能强大的工具,用于验证测试结果是否符合预期。它广泛用于自动化测试和 CI/CD 流水线,以确保代码的正确性。 - -## 激活时机 -- 当编写需要验证命令成功或失败的 Shell 脚本时。 -- 当对文件系统执行回归测试时(检查非空、存在等)。 -- 当验证变量状态或检测全局变量泄漏时。 -- 当检查 stdout/stderr 是否符合预期模式或内容时。 - -## 核心原则与规范 -- **关注退出码**: 命令成功时返回 0,失败时返回非 0,如果与 `set -e` 结合使用,可以中断脚本执行。 -- **支持否定**: 使用 `!` 进行命令的否定检查,或使用 `^` 进行类型/文件的否定检查。 -- **批量验证**: 许多子命令支持一次性检查多个值或文件。 - -## 补充场景 -- **变量快照**: 使用 `var save` 和 `var cmp` 检测函数中意外的全局变量修改。 -- **输出匹配**: 使用 `stdout` 结合 heredoc 验证复杂的、多行的命令输出。 - -## 实战示例 - -### 命令与类型验证 -```bash -# 验证命令成功以及多个整数值 -x assert true [ 1 -eq 1 ] -x assert is-int 10 20 30 -``` - -### 文件系统检查 -```bash -# 确保特定文件存在且非空 -x assert is-file /etc/hosts -x assert is-nonempty ./log.txt -``` - -### 检测变量泄漏 -```bash -# 捕获状态,运行函数,并检查泄漏 -x assert var save -my_function -x assert var cmp -``` - -## 交付验证清单 -- [ ] 确保使用了正确的操作符(`is-file`、`is-int` 等)。 -- [ ] 确认测试逻辑是否需要否定(`!` 或 `^`)。 -- [ ] 验证多行预期内容是否与 stdin/heredoc 格式匹配。 diff --git a/data/x-cmd/brew/SKILL.md b/data/x-cmd/brew/SKILL.md deleted file mode 100644 index 67cea23..0000000 --- a/data/x-cmd/brew/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: brew -description: > - Enhanced Homebrew wrapper for macOS and Linux, managing packages, mirrors, and proxies. - Core Scenario: When the user needs to install, manage, or configure Homebrew packages with optimized downloads. -license: MIT ---- - -# brew - Enhanced Homebrew Wrapper - -The `brew` module enhances the standard Homebrew experience by providing easier access to common tasks like mirror configuration, proxy setup, and interactive package selection. - -## When to Activate -- When installing, removing, or listing Homebrew packages on macOS or Linux. -- When configuring Homebrew to use optimized regional mirrors (e.g., TUNA). -- When managing Homebrew's proxy settings for restricted networks. -- When performing interactive package browsing (`x brew` app). - -## Core Principles & Rules -- **Interactive Browsing**: Use the default `x brew` to choose packages via TUI. -- **Mirror Support**: Simplify mirror switching with the `mirror` subcommand. -- **Privacy Focus**: Easily disable Homebrew analytics via the `analytics` subcommand. - -## Patterns & Examples - -### Install Packages -```bash -# Install multiple packages at once -x brew install curl wget git -``` - -### Configure Mirror -```bash -# Set Homebrew to use the Tsinghua University mirror -x brew mirror set tuna -``` - -### Interactive App -```bash -# Launch interactive TUI to manage Homebrew packages -x brew -``` - -## Checklist -- [ ] Confirm the names of the packages to be installed or managed. -- [ ] Verify if a mirror or proxy setup is needed for better connectivity. -- [ ] Check if the system is macOS or Linux (supported platforms). diff --git a/data/x-cmd/brew/SKILL.zh.md b/data/x-cmd/brew/SKILL.zh.md deleted file mode 100644 index 75da2de..0000000 --- a/data/x-cmd/brew/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: brew -description: > - 适用于 macOS 和 Linux 的增强型 Homebrew 封装,管理软件包、镜像和代理。 - 核心场景:当用户需要在优化下载的情况下安装、管理或配置 Homebrew 软件包时。 -license: MIT ---- - -# brew - 增强型 Homebrew 封装 - -`brew` 模块通过提供对镜像配置、代理设置和交互式包选择等常见任务的便捷访问,增强了标准的 Homebrew 体验。 - -## 激活时机 -- 在 macOS 或 Linux 上安装、移除或罗列 Homebrew 软件包时。 -- 当配置 Homebrew 使用优化的地区镜像(如 TUNA)时。 -- 当为受限网络管理 Homebrew 的代理设置时。 -- 当执行交互式软件包浏览 (`x brew` 应用) 时。 - -## 核心原则与规范 -- **交互式浏览**: 使用默认的 `x brew` 通过 TUI 选择软件包。 -- **镜像支持**: 使用 `mirror` 子命令简化镜像切换。 -- **隐私关注**: 通过 `analytics` 子命令轻松禁用 Homebrew 分析。 - -## 实战示例 - -### 安装软件包 -```bash -# 一次安装多个软件包 -x brew install curl wget git -``` - -### 配置镜像 -```bash -# 设置 Homebrew 使用清华大学镜像源 -x brew mirror set tuna -``` - -### 交互式应用 -```bash -# 启动交互式 TUI 来管理 Homebrew 软件包 -x brew -``` - -## 交付验证清单 -- [ ] 确认要安装或管理的软件包名称。 -- [ ] 验证是否需要设置镜像或代理以获得更好的连通性。 -- [ ] 检查系统是 macOS 还是 Linux(受支持的平台)。 diff --git a/data/x-cmd/bun/SKILL.md b/data/x-cmd/bun/SKILL.md deleted file mode 100644 index 0a0dc67..0000000 --- a/data/x-cmd/bun/SKILL.md +++ /dev/null @@ -1,34 +0,0 @@ ---- -name: bun -description: > - High-performance JavaScript runtime and package manager module. - Core Scenario: When the user needs an extremely fast JS/TS runtime or package manager for local development. -license: MIT ---- - -# bun - High-Performance JS/TS Tooling - -The `bun` module provides a streamlined interface for Bun, the all-in-one JavaScript runtime, packager, and test runner focused on speed. - -## When to Activate -- When requiring a high-speed runtime for JavaScript or TypeScript. -- When using Bun as a fast alternative to npm or yarn for package management. -- When performing local development tasks that benefit from Bun's performance. - -## Patterns & Examples - -### Run with Bun -```bash -# Execute a script using the Bun runtime -x bun run main.ts -``` - -### Fast Installation -```bash -# Use Bun to install project dependencies quickly -x bun install -``` - -## Checklist -- [ ] Confirm if the environment supports Bun. -- [ ] Verify if performance is a key requirement for the task. diff --git a/data/x-cmd/bun/SKILL.zh.md b/data/x-cmd/bun/SKILL.zh.md deleted file mode 100644 index 35b8ded..0000000 --- a/data/x-cmd/bun/SKILL.zh.md +++ /dev/null @@ -1,34 +0,0 @@ ---- -name: bun -description: > - 高性能 JavaScript 运行时和包管理器模块。 - 核心场景:当用户需要极速的 JS/TS 运行时或用于本地开发的包管理器时。 -license: MIT ---- - -# bun - 高性能 JS/TS 工具集 - -`bun` 模块为 Bun 提供了一个简化的界面,Bun 是一款专注于速度的全能 JavaScript 运行时、打包器和测试运行器。 - -## 激活时机 -- 当需要高性能的 JavaScript 或 TypeScript 运行时。 -- 当使用 Bun 作为 npm 或 yarn 的快速替代方案进行包管理时。 -- 当执行能从 Bun 的性能中获益的本地开发任务时。 - -## 实战示例 - -### 使用 Bun 运行 -```bash -# 使用 Bun 运行时执行脚本 -x bun run main.ts -``` - -### 快速安装 -```bash -# 使用 Bun 快速安装项目依赖项 -x bun install -``` - -## 交付验证清单 -- [ ] 确认环境是否支持 Bun。 -- [ ] 验证性能是否是该任务的关键要求。 diff --git a/data/x-cmd/buse/SKILL.md b/data/x-cmd/buse/SKILL.md deleted file mode 100644 index 91f98a0..0000000 --- a/data/x-cmd/buse/SKILL.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: buse -description: > - Enhancement module for browser-use, allowing AI to control browsers using natural language. - Core Scenario: When the user wants an AI agent to perform tasks in a browser (e.g., searching docs, taking screenshots) via terminal. -license: MIT ---- - -# buse - Natural Language Browser Automation - -The `buse` module enables users to control a web browser using natural language through the `browser-use` framework. It automates complex browser actions without requiring manual coding. - -## When to Activate -- When the user wants an AI to navigate websites or interact with web elements. -- When performing automated browser tasks like documentation search or visual verification (screenshots). -- When using a terminal UI to interactively guide a browser agent. -- When running headless browser tasks for data extraction or testing. - -## Core Principles & Rules -- **Safety & Permissions**: AI-generated browser actions should be monitored; use caution with sensitive sites. -- **Model Requirement**: Requires an AI model API key (OpenAI, Gemini, etc.) configured via respective x-cmd modules. -- **Context Handling**: Supports custom browser windows, user data directories, and CDP connections. - -## Additional Scenarios -- **Visual Evidence**: Automatically take screenshots during a browser task using specific prompts. -- **MCP Mode**: Run as a Model Context Protocol server for integration with other AI tools. - -## Patterns & Examples - -### Interactive Browser Control -```bash -# Launch the interactive terminal UI for browser control -x buse -``` - -### Direct Task Execution -```bash -# Run a specific browser task non-interactively -x buse -p "Search for OpenAI documentation and take a screenshot of the homepage" -``` - -### Headless Mode -```bash -# Run a task in the background without a visible window -x buse --headless -p "Check the stock status of an item on ExampleStore.com" -``` - -## Checklist -- [ ] Ensure browser-use and Chromium are installed via `x buse --install`. -- [ ] Confirm that at least one AI model API key is set up. -- [ ] Verify if the task requires a visible window or can run headless. diff --git a/data/x-cmd/buse/SKILL.zh.md b/data/x-cmd/buse/SKILL.zh.md deleted file mode 100644 index 083294e..0000000 --- a/data/x-cmd/buse/SKILL.zh.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: buse -description: > - browser-use 增强模块,允许 AI 使用自然语言控制浏览器。 - 核心场景:当用户希望 AI 代理通过终端在浏览器中执行任务(如搜索文档、截图)时。 -license: MIT ---- - -# buse - 自然语言浏览器自动化 - -`buse` 模块使用户能够通过 `browser-use` 框架,使用自然语言控制网络浏览器。它能自动执行复杂的浏览器操作,而无需手动编写代码。 - -## 激活时机 -- 当用户希望 AI 导航网站或与网页元素交互时。 -- 当执行自动化的浏览器任务(如文档搜索或视觉验证/截图)时。 -- 当使用终端 UI 交互式地引导浏览器代理时。 -- 当为数据提取或测试运行无头浏览器任务时。 - -## 核心原则与规范 -- **安全与权限**: AI 生成的浏览器操作应受到监控;在处理敏感网站时需谨慎。 -- **模型要求**: 需要通过相应的 x-cmd 模块配置 AI 模型 API 密钥(OpenAI, Gemini 等)。 -- **上下文处理**: 支持自定义浏览器窗口、用户数据目录和 CDP 连接。 - -## 补充场景 -- **视觉证据**: 在浏览器任务期间使用特定提示词自动截取屏幕截图。 -- **MCP 模式**: 作为模型上下文协议服务器运行,以便与其他 AI 工具集成。 - -## 实战示例 - -### 交互式浏览器控制 -```bash -# 启动用于浏览器控制的交互式终端 UI -x buse -``` - -### 直接任务执行 -```bash -# 以非交互方式运行特定的浏览器任务 -x buse -p "搜索 OpenAI 文档并截取首页屏幕截图" -``` - -### 无头模式 -```bash -# 在后台运行任务,不显示可见窗口 -x buse --headless -p "检查 ExampleStore.com 上某商品的库存状态" -``` - -## 交付验证清单 -- [ ] 确保已通过 `x buse --install` 安装了 browser-use 和 Chromium。 -- [ ] 确认至少设置了一个 AI 模型 API 密钥。 -- [ ] 验证任务是否需要可见窗口,或者可以在无头模式下运行。 diff --git a/data/x-cmd/bwh/SKILL.md b/data/x-cmd/bwh/SKILL.md deleted file mode 100644 index 8798a9c..0000000 --- a/data/x-cmd/bwh/SKILL.md +++ /dev/null @@ -1,55 +0,0 @@ ---- -name: bwh -description: > - BandwagonHost (BWH) VPS CLI management tool. - Core Scenario: When AI needs to manage VPS status, retrieve server info, or execute remote scripts. -license: MIT ---- - -# x bwh - VPS Management Assistant (AI Optimized) - -The `x bwh` module allows managing your BandwagonHost VPS via the command line. It's ideal for server status control, information retrieval, and remote command execution within AI workflows. - -## When to Activate -- When querying public IP, plan details, traffic limits, or real-time status of a VPS. -- When remote restarting, starting, or stopping the server. -- When retrieving SSH ports or resetting Root passwords. - -## Core Principles & Rules -- **Non-interactive First**: Directly use subcommands to get structured output. -- **Environment Requirements**: This module requires an API Key and VEID. If not configured, AI should guide the user through initialization. -- **Configuration Guidance**: - - Direct the user to the BandwagonHost portal to get API info. - - Suggest the user run `x bwh init` for configuration. - -## Patterns & Examples - -### Get VPS Detailed Info -```bash -# Get IP, traffic, expiry, and other info for the configured VPS -x bwh info -``` - -### Control Server Status -```bash -# Restart the server -x bwh restart - -# Start the server -x bwh start -``` - -### Query SSH Port -```bash -# Get the current SSH port, useful for subsequent SSH connections -x bwh info | grep "SSH Port" -``` - -## Configuration Guide (for AI) -If an `Unauthorized` error occurs or configuration is missing, provide this guidance to the user: -> Please obtain your API Key and VEID from the BandwagonHost website, then run the following command in your terminal to initialize: -> `x bwh init` - -## Checklist -- [ ] Confirm if API Key and VEID are configured. -- [ ] Confirm before performing destructive actions like `restart` or `reinstall`. diff --git a/data/x-cmd/bwh/SKILL.zh.md b/data/x-cmd/bwh/SKILL.zh.md deleted file mode 100644 index 30cf14e..0000000 --- a/data/x-cmd/bwh/SKILL.zh.md +++ /dev/null @@ -1,55 +0,0 @@ ---- -name: bwh -description: > - BandwagonHost (搬瓦工) VPS 命令行管理工具。 - 核心场景:当 AI 需要管理 VPS 状态、获取服务器信息或在远程机器上执行简单脚本时。 -license: MIT ---- - -# x bwh - VPS 管理助手 (AI 优化版) - -`x bwh` 模块允许通过命令行管理您的 BandwagonHost VPS 服务器。它非常适合在脚本或 AI 工作流中执行服务器启停、信息查询及远程命令。 - -## 激活时机 -- 当需要查询 VPS 的公网 IP、套餐详情、流量限制或实时状态时。 -- 当需要远程重启、启动或停止服务器时。 -- 当需要获取 SSH 端口或重置 Root 密码时。 - -## 核心原则与规范 -- **非交互优先**: 直接使用子命令获取结构化输出。 -- **环境要求**: 此模块需要 API Key 和 VEID。如果未配置,AI 应引导用户初始化。 -- **配置引导**: - - 引导用户访问 BandwagonHost 后台获取 API 信息。 - - 建议用户运行 `x bwh init` 进行配置。 - -## 实战示例 - -### 获取 VPS 详细信息 -```bash -# 获取当前配置的 VPS 的 IP、流量、到期时间等信息 -x bwh info -``` - -### 控制服务器状态 -```bash -# 重启服务器 -x bwh restart - -# 启动服务器 -x bwh start -``` - -### 查询 SSH 端口 -```bash -# 获取 VPS 当前的 SSH 端口,这对于后续使用 ssh 连接非常有用 -x bwh info | grep "SSH Port" -``` - -## 配置指南 (针对 AI) -如果遇到 `Unauthorized` 或找不到配置的错误,请向用户输出以下引导: -> 请先在 BandwagonHost 官网获取 API Key 和 VEID,然后在终端运行以下命令进行初始化: -> `x bwh init` - -## 交付验证清单 -- [ ] 确认 API Key 和 VEID 是否已配置。 -- [ ] 在执行 `restart` 或 `reinstall` 等破坏性操作前确认。 diff --git a/data/x-cmd/cal/SKILL.md b/data/x-cmd/cal/SKILL.md deleted file mode 100644 index 825428a..0000000 --- a/data/x-cmd/cal/SKILL.md +++ /dev/null @@ -1,36 +0,0 @@ ---- -name: cal -description: > - Standard Gregorian calendar tool with support for zodiac signs and weekly formats. - Core Scenario: When the user needs to view standard calendars or query zodiac signs by date. -license: MIT ---- - -# cal - Enhanced Gregorian Calendar - -The `cal` module is an enhanced version of the classic calendar command, supporting standard Gregorian views along with extras like zodiac sign lookups. - -## When to Activate -- When the user wants to see a standard monthly or yearly calendar. -- When querying the zodiac sign for a specific birth date. - -## Core Principles & Rules -- **Localization**: Automatically outputs zodiac names in the preferred language (e.g., "狮子座" vs "Leo"). -- **Simplicity**: Focuses on being a fast, drop-in replacement for `cal`. - -## Patterns & Examples - -### Query Zodiac -```bash -# Find the zodiac sign for a specific date -x cal zodiac 06-22 -``` - -### View Future Calendar -```bash -# Show the calendar for December 2024 -x cal 12 2024 -``` - -## Checklist -- [ ] Confirm if the user needs a calendar view or a zodiac query. diff --git a/data/x-cmd/cal/SKILL.zh.md b/data/x-cmd/cal/SKILL.zh.md deleted file mode 100644 index 6e63aa0..0000000 --- a/data/x-cmd/cal/SKILL.zh.md +++ /dev/null @@ -1,36 +0,0 @@ ---- -name: cal -description: > - 标准格里历日历工具,支持星座和周格式。 - 核心场景:当用户需要查看标准日历或通过日期查询星座时。 -license: MIT ---- - -# cal - 增强型格里历日历 - -`cal` 模块是经典日历命令的增强版本,支持标准的格里历视图以及星座查询等附加功能。 - -## 激活时机 -- 当用户想要查看标准的月历或年历时。 -- 当查询特定出生日期的星座时。 - -## 核心原则与规范 -- **本地化**: 自动以首选语言输出星座名称(例如“狮子座”对比“Leo”)。 -- **简洁性**: 专注于作为 `cal` 的快速、直接替代品。 - -## 实战示例 - -### 查询星座 -```bash -# 查找特定日期的星座 -x cal zodiac 06-22 -``` - -### 查看未来日历 -```bash -# 显示 2024 年 12 月的日历 -x cal 12 2024 -``` - -## 交付验证清单 -- [ ] 确认用户需要的是日历视图还是星座查询。 diff --git a/data/x-cmd/cb/SKILL.md b/data/x-cmd/cb/SKILL.md deleted file mode 100644 index be6018c..0000000 --- a/data/x-cmd/cb/SKILL.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: cb -description: > - CLI client for Codeberg, providing tools for repository management and collaboration. - Core Scenario: When the user needs to manage projects or collaborate on the Codeberg platform. -license: MIT ---- - -# cb - Codeberg CLI Browser - -The `cb` module provides a CLI interface for Codeberg, an open-source collaboration platform. It simplifies common tasks like repository listing and user management. - -## When to Activate -- When managing projects hosted on Codeberg. -- When performing rapid repository lookups or user profile checks on Codeberg. - -## Patterns & Examples - -### View Profile -```bash -# Display info for a specific Codeberg user -x cb user info username -``` - -## Checklist -- [ ] Confirm the Codeberg username or repository name. diff --git a/data/x-cmd/cb/SKILL.zh.md b/data/x-cmd/cb/SKILL.zh.md deleted file mode 100644 index c7581c4..0000000 --- a/data/x-cmd/cb/SKILL.zh.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: cb -description: > - Codeberg 的命令行客户端,提供用于仓库管理和协作的工具。 - 核心场景:当用户需要在 Codeberg 平台上管理项目或进行协作时。 -license: MIT ---- - -# cb - Codeberg 命令行浏览器 - -`cb` 模块为 Codeberg(一个开源协作平台)提供了命令行界面。它简化了如仓库罗列和用户管理等常见任务。 - -## 激活时机 -- 当管理托管在 Codeberg 上的项目时。 -- 当在 Codeberg 上执行快速的仓库查找或用户信息检查时。 - -## 实战示例 - -### 查看个人资料 -```bash -# 显示特定 Codeberg 用户的详细信息 -x cb user info username -``` - -## 交付验证清单 -- [ ] 确认 Codeberg 用户名或仓库名称。 diff --git a/data/x-cmd/ccal/SKILL.md b/data/x-cmd/ccal/SKILL.md deleted file mode 100644 index 57b87ec..0000000 --- a/data/x-cmd/ccal/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: ccal -description: > - Terminal calendar tool for the Chinese lunar calendar, solar terms, and traditional festivals. - Core Scenario: When the user needs to check lunar dates, solar terms, or traditional Chinese holiday information. -license: MIT ---- - -# ccal - Chinese Lunar Calendar - -The `ccal` module is a terminal calendar specifically designed for users interested in the Chinese lunar calendar. It provides lunar dates, solar terms, and information about traditional festivals and daily fortune (Huangli). - -## When to Activate -- When the user wants to see the current or a specific date's lunar info. -- When checking solar terms, public holidays, or makeup work days in China. -- When viewing traditional festival dates or zodiac information interactively. - -## Core Principles & Rules -- **Interactive Browsing**: Use `fz` subcommands for a searchable TUI experience. -- **Detailed Info**: Use the `info` subcommand for a comprehensive breakdown of a single day (lunar, stems-and-branches, etc.). -- **Localization**: Supports customizing weekday naming (e.g., using "礼拜" vs "星期"). - -## Patterns & Examples - -### Today's Lunar Info -```bash -# View detailed lunar and traditional info for today -x ccal info -``` - -### Specific Date -```bash -# Check lunar info for a future date -x ccal info 2025-5-20 -``` - -### Interactive Year View -```bash -# Browse traditional festivals and zodiacs interactively for the year -x ccal fzy -``` - -## Checklist -- [ ] Confirm if the user needs info for today or a specific date. -- [ ] Verify if an interactive view or a simple data output is preferred. diff --git a/data/x-cmd/ccal/SKILL.zh.md b/data/x-cmd/ccal/SKILL.zh.md deleted file mode 100644 index d50f934..0000000 --- a/data/x-cmd/ccal/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: ccal -description: > - 用于中国农历、节气和传统节日的终端日历工具。 - 核心场景:当用户需要检查农历日期、节气或中国传统节日信息时。 -license: MIT ---- - -# ccal - 中国农历 - -`ccal` 模块是专为对中国农历感兴趣的用户设计的终端日历。它提供农历日期、节气以及关于传统节日和每日黄历的信息。 - -## 激活时机 -- 当用户想要查看当前或特定日期的农历信息时。 -- 当检查中国的节气、公众假期或调休上班日期时。 -- 当交互式地查看传统节日日期或生肖信息时。 - -## 核心原则与规范 -- **交互式浏览**: 使用 `fz` 子命令获得可搜索的 TUI 体验。 -- **详细信息**: 使用 `info` 子命令获取单日的全面分解(农历、干支等)。 -- **本地化**: 支持自定义星期命名(例如使用“礼拜”与“星期”)。 - -## 实战示例 - -### 今日农历信息 -```bash -# 查看今天的详细农历和传统信息 -x ccal info -``` - -### 特定日期 -```bash -# 检查未来日期的农历信息 -x ccal info 2025-5-20 -``` - -### 交互式年视图 -```bash -# 交互式浏览全年的传统节日和生肖 -x ccal fzy -``` - -## 交付验证清单 -- [ ] 确认用户需要的是今天还是特定日期的信息。 -- [ ] 验证首选的是交互式视图还是简单的信息输出。 diff --git a/data/x-cmd/ccmd/SKILL.md b/data/x-cmd/ccmd/SKILL.md deleted file mode 100644 index 8f744a5..0000000 --- a/data/x-cmd/ccmd/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: ccmd -description: > - Cache command execution results to avoid repeated execution of time-consuming tasks. - Core Scenario: When the user needs to speed up workflows involving slow network requests (curl) or complex local queries. -license: MIT ---- - -# ccmd - Command Output Caching - -The `ccmd` module caches the stdout of commands for a specified duration, preventing unnecessary re-execution of slow or resource-intensive processes. - -## When to Activate -- When performing repeated network requests (e.g., `curl`). -- When running slow diagnostic tools or complex file system searches. -- When a user wants to maintain a "snapshot" of a command's output for a period. - -## Core Principles & Rules -- **Duration Format**: Supports time units like `s`, `m`, `h`, `d`, `w` (e.g., `1h` for one hour). Default is `1d`. -- **Command Separation**: The `--` separator is mandatory when providing the command to be executed. -- **Cache Management**: Use `invalidate` to force a refresh or `clear` to remove all cached data. - -## Patterns & Examples - -### Cache HTTP Request -```bash -# Cache the result of a curl command for 1 hour -x ccmd 1h -- curl https://api.example.com/data -``` - -### Cache Local Query -```bash -# Cache a geographical IP lookup for 30 minutes -x ccmd 30m -- x ips geo 1.1.1.1 -``` - -### Force Refresh -```bash -# Clear the cache for a specific command to ensure next run is live -x ccmd invalidate curl -``` - -## Checklist -- [ ] Confirm the desired cache duration. -- [ ] Ensure the `--` separator is correctly placed. -- [ ] Verify if the command output is suitable for caching (i.e., not real-time sensitive). diff --git a/data/x-cmd/ccmd/SKILL.zh.md b/data/x-cmd/ccmd/SKILL.zh.md deleted file mode 100644 index 8f91b25..0000000 --- a/data/x-cmd/ccmd/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: ccmd -description: > - 缓存命令执行结果,避免重复执行耗时任务。 - 核心场景:当用户需要加速涉及缓慢网络请求 (curl) 或复杂本地查询的工作流时。 -license: MIT ---- - -# ccmd - 命令输出缓存 - -`ccmd` 模块将命令的 stdout 缓存指定的时间,从而防止不必要地重复执行缓慢或资源密集型进程。 - -## 激活时机 -- 当执行重复的网络请求时(例如 `curl`)。 -- 当运行缓慢的诊断工具或复杂的分布式文件系统搜索时。 -- 当用户想要在一段时间内保持命令输出的“快照”时。 - -## 核心原则与规范 -- **时长格式**: 支持时间单位 `s`、`m`、`h`、`d`、`w`(例如 `1h` 表示一小时)。默认为 `1d`。 -- **命令分隔**: 在提供要执行的命令时,`--` 分隔符是强制性的。 -- **缓存管理**: 使用 `invalidate` 强制刷新,或使用 `clear` 删除所有缓存数据。 - -## 实战示例 - -### 缓存 HTTP 请求 -```bash -# 将 curl 命令的结果缓存 1 小时 -x ccmd 1h -- curl https://api.example.com/data -``` - -### 缓存本地查询 -```bash -# 将地理 IP 查询缓存 30 分钟 -x ccmd 30m -- x ips geo 1.1.1.1 -``` - -### 强制刷新 -```bash -# 清除特定命令的缓存以确保下次运行是实时的 -x ccmd invalidate curl -``` - -## 交付验证清单 -- [ ] 确认所需的缓存时长。 -- [ ] 确保 `--` 分隔符放置正确。 -- [ ] 验证命令输出是否适合缓存(即不是对实时性极度敏感的数据)。 diff --git a/data/x-cmd/cd/SKILL.md b/data/x-cmd/cd/SKILL.md deleted file mode 100644 index 3367cab..0000000 --- a/data/x-cmd/cd/SKILL.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: cd -description: > - Enhanced cd command with history navigation, fuzzy search, and directory aliases. - Core Scenario: When the user needs to quickly jump between historical paths or search for subdirectories using keywords. -license: MIT ---- - -# cd - Enhanced Directory Navigation - -The `cd` module is a powerful replacement for the standard `cd` command, implemented in shell and awk. It offers interactive history navigation, keyword-based jumping, and directory aliasing. - -## When to Activate -- When the user wants to jump to a directory based on a keyword in its path. -- When navigating back to a parent directory using a keyword match (`-b`). -- When performing a forward search for subdirectories (`-f`). -- When managing or using directory "realms" (aliases). -- When executing a command in another directory and returning immediately. - -## Core Principles & Rules -- **Seamless Jump**: Use `,` for interactive history search or `,keyword` for direct jumps. -- **Bi-directional Navigation**: Use `-b` to go "backwards" (up the tree) and `-f` to go "forwards" (down into subfolders). -- **Execution Wrappers**: Support for executing commands in a target directory without permanently changing the session path. - -## Patterns & Examples - -### Smart Jumping -```bash -# Open interactive history search -c -# Jump to a historical path matching 'project' -c ,project -``` - -### Contextual Navigation -```bash -# Go up the tree to the nearest directory matching 'src' -c -b src -# Find and jump into a subfolder matching 'test' -c -f test -``` - -### Temporary Execution -```bash -# Run 'ls -la' in /bin and then return automatically -c /bin ls -la -``` - -## Checklist -- [ ] Confirm if the jump is based on history (`,`) or parent/child relationship. -- [ ] Verify if the user intended to stay in the target directory or just run a command. -- [ ] Check if a directory alias (`:realm`) exists if used. diff --git a/data/x-cmd/cd/SKILL.zh.md b/data/x-cmd/cd/SKILL.zh.md deleted file mode 100644 index 2819b68..0000000 --- a/data/x-cmd/cd/SKILL.zh.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: cd -description: > - 增强型 cd 命令,具有历史导航、模糊搜索和目录别名功能。 - 核心场景:当用户需要快速在历史路径之间跳转或使用关键词搜索子目录时。 -license: MIT ---- - -# cd - 增强型目录导航 - -`cd` 模块是标准 `cd` 命令的强大替代方案,使用 Shell 和 awk 实现。它提供交互式历史导航、基于关键词的跳转以及目录别名功能。 - -## 激活时机 -- 当用户想要根据路径中的关键词跳转到目录时。 -- 当使用关键词匹配回退到上级目录时 (`-b`)。 -- 当对子目录执行前向搜索时 (`-f`)。 -- 当管理或使用目录“领域” (别名) 时。 -- 当在另一个目录执行命令并立即返回时。 - -## 核心原则与规范 -- **无缝跳转**: 使用 `,` 进行交互式历史搜索,或使用 `,关键词` 进行直接跳转。 -- **双向导航**: 使用 `-b` 向“后”跳转(向上级树),使用 `-f` 向“前”跳转(向下级子文件夹)。 -- **执行包装器**: 支持在目标目录中执行命令,而不会永久更改会话路径。 - -## 实战示例 - -### 智能跳转 -```bash -# 打开交互式历史搜索 -c -# 跳转到匹配 'project' 的历史路径 -c ,project -``` - -### 上下文导航 -```bash -# 向上跳转到最近的匹配 'src' 的上级目录 -c -b src -# 查找并跳转到匹配 'test' 的子文件夹 -c -f test -``` - -### 临时执行 -```bash -# 在 /bin 中运行 'ls -la' 然后自动返回 -c /bin ls -la -``` - -## 交付验证清单 -- [ ] 确认跳转是基于历史 (`,`) 还是父子关系。 -- [ ] 验证用户是打算留在目标目录还是仅仅运行一条命令。 -- [ ] 如果使用了目录别名 (`:realm`),请检查其是否存在。 diff --git a/data/x-cmd/cht/SKILL.md b/data/x-cmd/cht/SKILL.md deleted file mode 100644 index 588a4d8..0000000 --- a/data/x-cmd/cht/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: cht -description: > - CLI client for cheat.sh, providing concise "cheat sheets" and code snippets for developers. - Core Scenario: When the user needs short, practical examples for programming languages or CLI tools. -license: MIT ---- - -# cht - Developer Cheat Sheets & Snippets - -The `cht` module leverages the cheat.sh platform to provide high-quality, community-vetted code snippets and command usage examples. It is designed for developers who need "how-to" answers instantly. - -## When to Activate -- When the user needs a code example for a specific programming language (e.g., `python/hello`). -- When seeking a concise "cheat sheet" for a command (e.g., `ls`). -- When the user wants to learn a language quickly using "learn-x-in-minutes" content (`--learn`). -- When searching across multiple cheat sheets using keywords. - -## Core Principles & Rules -- **Example-Driven**: Focus on providing runnable code snippets or direct command examples. -- **Breadth of Content**: Covers hundreds of CLI tools and dozens of programming languages. -- **Search Capability**: Use `-s` for keyword-based search within the cheat.sh ecosystem. - -## Patterns & Examples - -### Language Snippet -```bash -# Get a Python 'hello world' example from cheat.sh -x cht python/hello -``` - -### Command Cheat Sheet -```bash -# View a concise summary of common 'sed' patterns -x cht sed -``` - -### Quick Learning -```bash -# Display the beginner cheat sheet for the Go language -x cht --learn go -``` - -## Checklist -- [ ] Confirm if the user needs a command reference or a language snippet. -- [ ] Verify if a specific task (like `hello`) should be appended to the language query. diff --git a/data/x-cmd/cht/SKILL.zh.md b/data/x-cmd/cht/SKILL.zh.md deleted file mode 100644 index 6a5759c..0000000 --- a/data/x-cmd/cht/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: cht -description: > - cheat.sh 的命令行客户端,为开发者提供简洁的“速查表”和代码片段。 - 核心场景:当用户需要编程语言或 CLI 工具的简短、实用的示例时。 -license: MIT ---- - -# cht - 开发者速查表与代码片段 - -`cht` 模块利用 cheat.sh 平台提供高质量的、经社区验证的代码片段和命令使用示例。它专为需要即时获取“如何做”答案的开发者而设计。 - -## 激活时机 -- 当用户需要特定编程语言的代码示例时(例如 `python/hello`)。 -- 当寻求命令的简洁“速查表”时(例如 `ls`)。 -- 当用户想要使用 “learn-x-in-minutes” 内容快速学习一门语言时 (`--learn`)。 -- 当使用关键词在多个速查表中进行搜索时。 - -## 核心原则与规范 -- **示例驱动**: 专注于提供可运行的代码片段或直接的命令示例。 -- **内容广度**: 涵盖数百种 CLI 工具和数十种编程语言。 -- **搜索能力**: 使用 `-s` 在 cheat.sh 生态系统中进行基于关键词的搜索。 - -## 实战示例 - -### 语言代码片段 -```bash -# 从 cheat.sh 获取 Python 的 'hello world' 示例 -x cht python/hello -``` - -### 命令速查表 -```bash -# 查看 'sed' 常用模式的简洁摘要 -x cht sed -``` - -### 快速学习 -```bash -# 显示 Go 语言的初学者速查表 -x cht --learn go -``` - -## 交付验证清单 -- [ ] 确认用户需要的是命令参考还是语言代码片段。 -- [ ] 验证是否应将特定任务(如 `hello`)附加到语言查询中。 diff --git a/data/x-cmd/claude/SKILL.md b/data/x-cmd/claude/SKILL.md deleted file mode 100644 index 4298c5f..0000000 --- a/data/x-cmd/claude/SKILL.md +++ /dev/null @@ -1,69 +0,0 @@ ---- -name: claude -description: > - Enhanced CLI module for Claude Code, supporting third-party model providers and session management. - Core Scenario: When the user wants to use Claude Code with alternative models (DeepSeek, etc.), track token usage, or manage MCP servers. -license: MIT ---- - -# claude - Claude Code Enhancement Module - -The `claude` module extends the capabilities of Anthropic's `claude-code`, allowing users to connect to various AI model providers, manage sessions, and monitor token usage through a unified CLI. - -## When to Activate -- When the user wants to start a Claude Code session with a specific provider (e.g., DeepSeek, Kimi, Moonshot, SiliconFlow, Doubao, MiniMax, Zhipu, etc.). -- When the user needs to configure Claude Code to use a third-party model globally or per project. -- When the user wants to track token consumption and costs for the last 7 days. -- When the user needs to manage Model Context Protocol (MCP) servers. -- When the user wants to clean up or customize Claude Code's status line or author attribution. - -## Core Principles & Rules -- **Provider Connection**: Use subcommands like `ds` (DeepSeek), `kimi`, `moonshot`, `sili` (SiliconFlow), `doubao`, `mm` (MiniMax), `zhipu` (GLM), `or` (OpenRouter), `ali` (Qwen), `baidu` (Ernie), `tencent` (Hunyuan), or `mt` (Mthreads) to launch Claude Code with those specific backends. -- **Global vs. Project Config**: Use `use` for global settings and `use --project` for project-specific model overrides. -- **Session Management**: Use `sess` (Interactive FZF App) or `resume` to manage and recover previous conversations. -- **Security**: Be cautious when using the `--dangerously-skip-permissions` flag; it should only be recommended in trusted sandbox environments. - -## Additional Scenarios -- **Local LLM**: Connect Claude Code to a local Ollama service using `x claude use other`. -- **Token Analysis**: Run `x claude usage` to get a detailed breakdown of costs and token count. -- **Clean Git History**: Remove the automatic Co-Author signature from Git commits using `x claude attribution rm`. - -## Patterns & Examples - -### Launch with Third-party Providers -```bash -# Start Claude Code using the DeepSeek model -x claude ds - -# Start Claude Code using SiliconFlow -x claude sili - -# Start Claude Code using Kimi -x claude kimi -``` - -### Configure Project Model -```bash -# Set the current project to use Zhipu (GLM) model provider -x claude use --project glm - -# Set the current project to use SiliconFlow -x claude use --project sili -``` - -### Track Usage -```bash -# Analyze token usage and costs for the past 7 days -x claude usage -``` - -### Interactive Session Management -```bash -# Browse and manage local sessions via FZF -x claude sess -``` - -## Checklist -- [ ] Confirm if the user wants to use a specific model provider (DeepSeek, MiniMax, etc.). -- [ ] Verify if the configuration should be global or project-specific. -- [ ] Ensure the relevant API Key is configured for the chosen provider. diff --git a/data/x-cmd/claude/SKILL.zh.md b/data/x-cmd/claude/SKILL.zh.md deleted file mode 100644 index fe702d1..0000000 --- a/data/x-cmd/claude/SKILL.zh.md +++ /dev/null @@ -1,60 +0,0 @@ ---- -name: claude -description: > - Claude Code 命令行增强模块,支持第三方模型提供商接入和会话管理。 - 核心场景:当用户想要配合其他模型(如 DeepSeek)使用 Claude Code、统计 Token 消耗或管理 MCP 时。 -license: MIT ---- - -# claude - Claude Code 增强模块 - -`claude` 模块扩展了 Anthropic 的 `claude-code` 能力,允许用户连接多种 AI 模型提供商、管理会话并监控 Token 消耗。 - -## 激活时机 -- 当用户想要使用特定提供商(如 DeepSeek, MiniMax, 智谱)启动 Claude Code 时。 -- 当用户需要全局或按项目配置 Claude Code 使用第三方模型时。 -- 当用户想要查看过去 7 天的 Token 消耗和成本统计时。 -- 当用户需要管理模型上下文协议 (MCP) 服务器时。 -- 当用户想要移除或自定义 Claude Code 的作者署名或状态栏时。 - -## 核心原则与规范 -- **连接提供商**: 使用 `ds` (DeepSeek), `mm` (MiniMax), `zhipu` 或 `or` (OpenRouter) 等子命令启动 Claude Code。 -- **全局 vs 项目配置**: 使用 `use` 进行全局设置,使用 `use --project` 进行项目级模型覆盖。 -- **会话管理**: 使用 `sess` (交互式 FZF App) 或 `resume` 管理并恢复之前的对话。 -- **安全性**: 谨慎建议使用 `--dangerously-skip-permissions` 标志;仅在可信的沙盒环境中推荐。 - -## 补充场景 -- **本地 LLM**: 通过 `x claude use other` 将 Claude Code 连接到本地 Ollama 服务。 -- **Token 分析**: 运行 `x claude usage` 获取详细的成本和 Token 统计。 -- **清洁 Git 历史**: 使用 `x claude attribution rm` 移除 Git 提交时自动生成的 Co-Author 签名。 - -## 实战示例 - -### 使用 DeepSeek 启动 -```bash -# 使用 DeepSeek 模型启动 Claude Code -x claude ds -``` - -### 配置项目模型 -```bash -# 将当前项目配置为使用智谱 (GLM) 模型提供商 -x claude use --project glm -``` - -### 统计使用情况 -```bash -# 分析过去 7 天的 Token 使用情况和成本 -x claude usage -``` - -### 交互式会话管理 -```bash -# 通过 FZF 浏览并管理本地会话 -x claude sess -``` - -## 交付验证清单 -- [ ] 确认用户是否想要使用特定的模型提供商(DeepSeek, MiniMax 等)。 -- [ ] 验证配置应该是全局性的还是项目级的。 -- [ ] 确保已为所选提供商配置了相关的 API Key。 diff --git a/data/x-cmd/codex/SKILL.md b/data/x-cmd/codex/SKILL.md deleted file mode 100644 index 01aa62f..0000000 --- a/data/x-cmd/codex/SKILL.md +++ /dev/null @@ -1,63 +0,0 @@ ---- -name: codex -description: > - Enhanced CLI for OpenAI's Codex terminal agent, supporting local code analysis, sandboxing, and third-party models. - Core Scenario: When the user wants to use Codex with specific models (DeepSeek, etc.), enable sandboxed execution, or run natural language commands. -license: MIT ---- - -# codex - AI Code Assistant & Terminal Agent - -The `codex` module enhances OpenAI's `codex` terminal agent, enabling semantic code search, automated patch generation, and connection to various AI model providers within a secure, sandboxed environment. - -## When to Activate -- When the user wants to start a Codex session with a specific provider (e.g., DeepSeek, Kimi, Zhipu, Doubao, SiliconFlow). -- When the user needs to execute shell commands or generate code based on natural language descriptions. -- When the user requires a sandboxed execution environment (`read-only`, `workspace-write`) for safety. -- When the user wants to apply generated diffs to their Git worktree. -- When the user needs to inject specific "skills" into the Codex agent's environment. - -## Core Principles & Rules -- **Sandboxing**: Always encourage using appropriate sandbox strategies (`--sandbox`) to prevent unintended system changes. -- **Automation Levels**: Use `--full-auto` for a balance between speed and safety (approves on failure). -- **Provider Switching**: Use subcommands like `ds` (DeepSeek), `kimi`, `zhipu`, `doubao`, or `sili` (SiliconFlow) to launch Codex with these providers. -- **Non-Interactive Execution**: Use `exec` or `e` for quick, one-off commands or code generation tasks. - -## Additional Scenarios -- **Web Search**: Enable real-time web search capabilities using the `--search` flag for up-to-date information. -- **Git Integration**: Quickly apply the latest generated diff from the agent using `x codex apply`. -- **Local OSS Models**: Connect to local Ollama services by using the `--oss` flag. - -## Patterns & Examples - -### Run Natural Language Command -```bash -# Execute a command described in English -x codex e "List all files larger than 10MB in the current directory" -``` - -### Start with Third-party Providers -```bash -# Launch Codex using DeepSeek with read-only sandbox protection -x codex ds --sandbox read-only - -# Launch Codex using SiliconFlow -x codex sili -``` - -### Apply Latest Patch -```bash -# Apply the latest generated code patch to the current Git repository -x codex apply -``` - -### Inject Skills -```bash -# Manage and inject custom skills into the Codex agent environment -x codex skill -``` - -## Checklist -- [ ] Confirm the desired sandbox level (`read-only`, `workspace-write`, etc.). -- [ ] Verify if a specific model provider (DeepSeek, Kimi, etc.) is needed. -- [ ] Ensure the user is aware of the risks when using `dangerously-bypass-approvals-and-sandbox`. diff --git a/data/x-cmd/codex/SKILL.zh.md b/data/x-cmd/codex/SKILL.zh.md deleted file mode 100644 index cd68002..0000000 --- a/data/x-cmd/codex/SKILL.zh.md +++ /dev/null @@ -1,60 +0,0 @@ ---- -name: codex -description: > - OpenAI Codex 终端代理增强命令行,支持本地代码库分析、沙箱运行及第三方模型。 - 核心场景:当用户想要配合特定模型(如 DeepSeek)使用 Codex、开启沙箱保护或执行自然语言命令时。 -license: MIT ---- - -# codex - AI 代码助手与终端代理 - -`codex` 模块增强了 OpenAI 的 `codex` 终端代理,支持语义代码搜索、自动补全、生成代码补丁,并在安全的沙箱环境中连接各种 AI 模型提供商。 - -## 激活时机 -- 当用户想要使用特定提供商(如 DeepSeek, Kimi, 智谱)启动 Codex 会话时。 -- 当用户需要根据自然语言描述执行 Shell 命令或生成代码时。 -- 当用户出于安全考虑需要沙箱运行环境(`read-only`, `workspace-write`)时。 -- 当用户想要将代理生成的 diff 应用到 Git 工作树时。 -- 当用户需要将特定“技能”(Skill) 注入到 Codex 代理环境时。 - -## 核心原则与规范 -- **沙箱保护**: 始终鼓励使用合适的沙箱策略 (`--sandbox`) 以防止意外的系统变更。 -- **自动化级别**: 使用 `--full-auto` 在速度和安全性之间取得平衡(仅在失败时请求确认)。 -- **模型提供商切换**: 使用 `ds` (DeepSeek), `kimi` 或 `zhipu` 等子命令启动 Codex。 -- **非交互式执行**: 使用 `exec` 或 `e` 快速完成单次命令执行或代码生成任务。 - -## 补充场景 -- **网页搜索**: 通过 `--search` 标志启用实时网页搜索能力,获取最新信息。 -- **Git 集成**: 使用 `x codex apply` 快速应用代理生成的最新代码补丁。 -- **本地开源模型**: 通过 `--oss` 标志连接到本地 Ollama 服务。 - -## 实战示例 - -### 执行自然语言命令 -```bash -# 执行用自然语言描述的命令 -x codex e "列出当前目录下大于 10MB 的所有文件" -``` - -### 在沙箱中使用 DeepSeek 启动 -```bash -# 使用 DeepSeek 并开启只读沙箱保护启动 Codex -x codex ds --sandbox read-only -``` - -### 应用最新补丁 -```bash -# 将代理生成的最新补丁应用到当前 Git 仓库 -x codex apply -``` - -### 注入 Skill -```bash -# 管理并注入自定义 Skill 到 Codex 代理环境 -x codex skill -``` - -## 交付验证清单 -- [ ] 确认所需的沙箱级别(`read-only`, `workspace-write` 等)。 -- [ ] 验证是否需要特定的模型提供商(DeepSeek, Kimi 等)。 -- [ ] 确保用户知晓使用 `dangerously-bypass-approvals-and-sandbox` 的风险。 diff --git a/data/x-cmd/coincap/SKILL.md b/data/x-cmd/coincap/SKILL.md deleted file mode 100644 index 3f5f17d..0000000 --- a/data/x-cmd/coincap/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: coincap -description: > - CLI viewer for CoinCap cryptocurrency market data, supporting real-time prices and structured exports. - Core Scenario: When the user needs current crypto prices, market caps, or data for automated financial analysis. -license: MIT ---- - -# coincap - Cryptocurrency Market Data - -The `coincap` module integrates the CoinCap API into the terminal, providing real-time data for over 1,000 cryptocurrencies. It supports interactive browsing and data export in JSON or CSV formats. - -## When to Activate -- When the user wants to check current cryptocurrency prices and market stats. -- When performing automated financial analysis using structured crypto data. -- When an interactive overview of the crypto market is required in the terminal. - -## Core Principles & Rules -- **API Key Required**: Remind users to obtain an API key for high-rate usage. -- **Data Portability**: Prioritize `--json` and `--csv` for script-based market analysis. -- **Interactive TUI**: Use `--app` for a visual, table-based browsing experience. - -## Patterns & Examples - -### View Market (Interactive) -```bash -# Open interactive TUI for real-time crypto prices -x coincap --app -``` - -### Export Market Data -```bash -# Get all market data as a JSON object -x coincap --json -``` - -### Initialize Config -```bash -# Set up API keys and session defaults -x coincap init -``` - -## Checklist -- [ ] Confirm if the user has an API key for prolonged use. -- [ ] Verify the desired output format (TUI, JSON, CSV). -- [ ] Ensure the user is aware the data is provided by CoinCap.io. diff --git a/data/x-cmd/coincap/SKILL.zh.md b/data/x-cmd/coincap/SKILL.zh.md deleted file mode 100644 index ff38f26..0000000 --- a/data/x-cmd/coincap/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: coincap -description: > - CoinCap 加密货币市场数据命令行查看器,支持实时价格和结构化导出。 - 核心场景:当用户需要当前的加密货币价格、市值或用于自动财务分析的数据时。 -license: MIT ---- - -# coincap - 加密货币市场数据 - -`coincap` 模块将 CoinCap API 集成到终端中,为 1,000 多种加密货币提供实时数据。它支持交互式浏览以及 JSON 或 CSV 格式的数据导出。 - -## 激活时机 -- 当用户想要检查当前的加密货币价格和市场统计数据时。 -- 当使用结构化加密数据执行自动化财务分析时。 -- 当终端中需要加密市场的交互式概览时。 - -## 核心原则与规范 -- **需要 API 密钥**: 提醒用户获取 API 密钥以进行高频率使用。 -- **数据便携性**: 为基于脚本的市场分析优先使用 `--json` 和 `--csv`。 -- **交互式 TUI**: 使用 `--app` 获得视觉化的表格浏览体验。 - -## 实战示例 - -### 查看市场 (交互式) -```bash -# 为实时加密货币价格打开交互式 TUI -x coincap --app -``` - -### 导出市场数据 -```bash -# 以 JSON 对象形式获取所有市场数据 -x coincap --json -``` - -### 初始化配置 -```bash -# 设置 API 密钥和会话默认值 -x coincap init -``` - -## 交付验证清单 -- [ ] 确认用户是否拥有用于长期使用的 API 密钥。 -- [ ] 验证所需的输出格式(TUI, JSON, CSV)。 -- [ ] 确保用户知晓数据由 CoinCap.io 提供。 diff --git a/data/x-cmd/cpu/SKILL.md b/data/x-cmd/cpu/SKILL.md deleted file mode 100644 index c4c8046..0000000 --- a/data/x-cmd/cpu/SKILL.md +++ /dev/null @@ -1,36 +0,0 @@ ---- -name: cpu -description: > - Retrieve CPU hardware information and detect system endianness. - Core Scenario: When the user needs to check processor details or identify byte order (endianness) for compilation or diagnostics. -license: MIT ---- - -# cpu - CPU Information & Detection - -The `cpu` module provides a quick way to retrieve detailed hardware information about the system's processor and determine the system's byte order (endianness). - -## When to Activate -- When identifying CPU architecture or hardware specs. -- When detecting system endianness (little-endian vs big-endian) for low-level development or network programming. - -## Core Principles & Rules -- **Conciseness**: Designed for rapid hardware checks. -- **Endianness Flags**: Returns `l` for little-endian and `b` for big-endian. - -## Patterns & Examples - -### View CPU Info -```bash -# Display detailed processor hardware information -x cpu info -``` - -### Detect Endianness -```bash -# Check system byte order -x cpu endianness -``` - -## Checklist -- [ ] Confirm if general info or just endianness is needed. diff --git a/data/x-cmd/cpu/SKILL.zh.md b/data/x-cmd/cpu/SKILL.zh.md deleted file mode 100644 index 7c0a4c4..0000000 --- a/data/x-cmd/cpu/SKILL.zh.md +++ /dev/null @@ -1,36 +0,0 @@ ---- -name: cpu -description: > - 获取 CPU 硬件信息并检测系统字节序。 - 核心场景:当用户需要检查处理器详情或为编译、诊断识别字节序时。 -license: MIT ---- - -# cpu - CPU 信息与检测 - -`cpu` 模块提供了一种快速获取系统处理器详细硬件信息并确定系统字节序的方法。 - -## 激活时机 -- 当识别 CPU 架构或硬件规格时。 -- 当为底层开发或网络编程检测系统字节序(小端 vs 大端)时。 - -## 核心原则与规范 -- **简洁性**: 旨在进行快速的硬件检查。 -- **字节序标志**: 小端返回 `l`,大端返回 `b`。 - -## 实战示例 - -### 查看 CPU 信息 -```bash -# 显示详细的处理器硬件信息 -x cpu info -``` - -### 检测字节序 -```bash -# 检查系统字节序 -x cpu endianness -``` - -## 交付验证清单 -- [ ] 确认需要的是常规信息还是仅字节序。 diff --git a/data/x-cmd/crush/SKILL.md b/data/x-cmd/crush/SKILL.md deleted file mode 100644 index 9d896ad..0000000 --- a/data/x-cmd/crush/SKILL.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: crush -description: > - AI-powered CLI tool by Charm for code analysis, diagnostics, and interactive Q&A. - Core Scenario: When the user needs intelligent code suggestions, explanations, or fast local development diagnostics. -license: MIT ---- - -# crush - AI Assistant for Code Analysis - -`crush` is an AI-driven tool developed by Charm that provides context-aware suggestions and code explanations, integrating with Language Server Protocol (LSP) for a smooth developer experience. - -## When to Activate -- When the user wants interactive AI assistance for code analysis. -- When performing fast diagnostics or seeking intelligent suggestions within a project. -- When running single, non-interactive AI prompts for code explanation. -- When the user needs a debugged or YOLO-approved execution of AI commands. - -## Core Principles & Rules -- **API Key Inheritance**: Inherits `OPENAI_API_KEY` and `GEMINI_API_KEY` from `x openai` and `x gemini` configurations. -- **Context Awareness**: Best used within a specific project directory (`--cwd`) to leverage local code context. -- **YOLO Mode**: Use `-y` or `--yolo` to automatically accept permissions for high-speed interaction. - -## Additional Scenarios -- **Non-Interactive Prompts**: Use the `run` subcommand for quick queries. -- **Debugging**: Enable detailed logs with `-d` for troubleshooting AI interactions. - -## Patterns & Examples - -### Interactive Session -```bash -# Start an interactive AI session for the current project -x crush -``` - -### Non-Interactive Query -```bash -# Ask for a quick code explanation -x crush run "Explain the use of context in Go" -``` - -### Sandboxed/CWD Execution -```bash -# Run crush in a specific project directory with debug logging -x crush -d -c /path/to/project -``` - -## Checklist -- [ ] Ensure relevant AI API keys (OpenAI/Gemini) are set up via their respective x-cmd modules. -- [ ] Verify the target directory if using the `--cwd` flag. -- [ ] Confirm if YOLO mode is appropriate for the current task safety. diff --git a/data/x-cmd/crush/SKILL.zh.md b/data/x-cmd/crush/SKILL.zh.md deleted file mode 100644 index 87cd0cd..0000000 --- a/data/x-cmd/crush/SKILL.zh.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: crush -description: > - 由 Charm 开发的 AI 驱动命令行工具,用于代码分析、诊断和交互式问答。 - 核心场景:当用户需要智能代码建议、解释或快速的本地开发诊断时。 -license: MIT ---- - -# crush - 用于代码分析的 AI 助手 - -`crush` 是由 Charm 开发的 AI 驱动工具,提供上下文感知的建议和代码解释,并与语言服务器协议 (LSP) 集成,提供流畅的开发体验。 - -## 激活时机 -- 当用户想要交互式的 AI 辅助进行代码分析时。 -- 当在项目内进行快速诊断或寻求智能建议时。 -- 当运行单个、非交互式的 AI 提示词进行代码解释时。 -- 当用户需要调试或以 YOLO 批准模式执行 AI 命令时。 - -## 核心原则与规范 -- **API 密钥继承**: 从 `x openai` 和 `x gemini` 配置中继承 `OPENAI_API_KEY` 和 `GEMINI_API_KEY`。 -- **上下文感知**: 最好在特定项目目录 (`--cwd`) 中使用,以利用本地代码上下文。 -- **YOLO 模式**: 使用 `-y` 或 `--yolo` 自动接受权限,以进行高速交互。 - -## 补充场景 -- **非交互式提示**: 使用 `run` 子命令进行快速查询。 -- **调试**: 使用 `-d` 启用详细日志,以排除 AI 交互故障。 - -## 实战示例 - -### 交互式会话 -```bash -# 为当前项目启动交互式 AI 会话 -x crush -``` - -### 非交互式查询 -```bash -# 请求快速的代码解释 -x crush run "解释 Go 语言中 context 的用法" -``` - -### 指定工作目录并开启调试 -```bash -# 在特定项目目录中运行 crush 并开启调试日志 -x crush -d -c /path/to/project -``` - -## 交付验证清单 -- [ ] 确保已通过相应的 x-cmd 模块设置了相关的 AI API 密钥 (OpenAI/Gemini)。 -- [ ] 如果使用 `--cwd` 标志,请验证目标目录。 -- [ ] 确认 YOLO 模式是否符合当前任务的安全性要求。 diff --git a/data/x-cmd/ddgo/SKILL.md b/data/x-cmd/ddgo/SKILL.md deleted file mode 100644 index 71945cf..0000000 --- a/data/x-cmd/ddgo/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: ddgo -description: > - CLI tool for DuckDuckGo search, supporting result summaries and AI-driven answer extraction. - Core Scenario: When the user needs to perform web searches or get AI-summarized answers from DuckDuckGo results. -license: MIT ---- - -# ddgo - DuckDuckGo CLI Search - -The `ddgo` module provides a powerful terminal-based interface for DuckDuckGo, enabling users to perform web searches, extract results in structured formats, and leverage AI for summarizing answers. - -## When to Activate -- When the user wants to perform a privacy-focused web search via CLI. -- When needing an AI-generated summary of search results for a specific query. -- When exporting search results to JSON for further processing. -- When performing site-specific searches (e.g., `site:x-cmd.com`). - -## Core Principles & Rules -- **AI Integration**: Use the `--ai` flag to automatically select and summarize the most relevant search result. -- **Data-Friendly**: Supports structured JSON output for script integration. -- **Structured Viewing**: Use `dump --app` for an interactive table view of search results. - -## Patterns & Examples - -### AI-Summarized Search -```bash -# Get an AI summary of search results for 'bash tips' -x ddgo --ai "bash best practices" -``` - -### Site-Specific Search -```bash -# Search for 'jq' content on the x-cmd.com website -x ddgo dump --json "site:x-cmd.com jq" -``` - -### Top Results -```bash -# Retrieve the top 10 search results for a query -x ddgo --top 10 "python threading tutorial" -``` - -## Checklist -- [ ] Confirm if the user needs a general search or an AI-summarized answer. -- [ ] Verify if the results should be limited to a specific site or number. -- [ ] Check if JSON output is required for subsequent tasks. diff --git a/data/x-cmd/ddgo/SKILL.zh.md b/data/x-cmd/ddgo/SKILL.zh.md deleted file mode 100644 index 1330ed0..0000000 --- a/data/x-cmd/ddgo/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: ddgo -description: > - DuckDuckGo 搜索的命令行工具,支持结果总结和 AI 驱动的答案提取。 - 核心场景:当用户需要执行网页搜索或从 DuckDuckGo 结果中获取 AI 总结的答案时。 -license: MIT ---- - -# ddgo - DuckDuckGo 命令行搜索 - -`ddgo` 模块为 DuckDuckGo 提供了一个强大的基于终端的界面,使用户能够执行网页搜索、以结构化格式提取结果,并利用 AI 总结答案。 - -## 激活时机 -- 当用户想要通过 CLI 执行注重隐私的网页搜索时。 -- 当针对特定查询需要 AI 生成的搜索结果总结时。 -- 当为进一步处理导出搜索结果到 JSON 时。 -- 当执行特定站点的搜索时(例如 `site:x-cmd.com`)。 - -## 核心原则与规范 -- **AI 集成**: 使用 `--ai` 标志自动选择并总结最相关的搜索结果。 -- **数据友好**: 支持结构化 JSON 输出以便脚本集成。 -- **结构化查看**: 使用 `dump --app` 以交互式表格视图查看搜索结果。 - -## 实战示例 - -### AI 总结搜索 -```bash -# 获取关于 'bash tips' 搜索结果的 AI 总结 -x ddgo --ai "bash best practices" -``` - -### 特定站点搜索 -```bash -# 在 x-cmd.com 网站上搜索 'jq' 内容 -x ddgo dump --json "site:x-cmd.com jq" -``` - -### 前几条结果 -```bash -# 检索查询的前 10 条搜索结果 -x ddgo --top 10 "python threading tutorial" -``` - -## 交付验证清单 -- [ ] 确认用户需要的是一般搜索还是 AI 总结的回答。 -- [ ] 验证结果是否应限制在特定的站点或数量内。 -- [ ] 检查后续任务是否需要 JSON 输出。 diff --git a/data/x-cmd/deepseek/SKILL.md b/data/x-cmd/deepseek/SKILL.md deleted file mode 100644 index 8f02de1..0000000 --- a/data/x-cmd/deepseek/SKILL.md +++ /dev/null @@ -1,49 +0,0 @@ ---- -name: deepseek -description: > - Integrate DeepSeek AI for high-performance text generation and specialized reasoning. - Core Scenario: When the user wants to use DeepSeek's V3 or Reasoner models for coding, translation, or complex logic. -license: MIT ---- - -# deepseek - DeepSeek AI Integration - -The `deepseek` module provides a CLI interface for DeepSeek's AI models, including the powerful V3 model and the reasoning-focused model. - -## When to Activate -- When the user wants to use DeepSeek for chat, text generation, or coding tasks. -- When the user needs complex reasoning or math solving using the `deepseek-reasoner` model. -- When checking DeepSeek account balance or managing API keys. - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` for setup. -- **Model Selection**: Default is V3 (`@ds`). Use `--model deepseek-reasoner` for tasks requiring deep thought. -- **Balance Monitoring**: Use the `balance` subcommand to keep track of credits. - -## Additional Scenarios -- **Quick Alias**: Use `@ds` for rapid interaction. -- **Project Configuration**: Set default models per project using the `--cfg` options. - -## Patterns & Examples - -### Chat with DeepSeek V3 -```bash -# Ask a general question -@ds "Explain the concept of quantum entanglement" -``` - -### Logical Reasoning -```bash -# Use the reasoning model for complex logic -@ds --model deepseek-reasoner "Solve this complex probability problem: ..." -``` - -### Check Account Balance -```bash -# View current credits -x deepseek balance -``` - -## Checklist -- [ ] Ensure the DeepSeek API key is initialized. -- [ ] Confirm if the reasoning model is better suited for the current task. diff --git a/data/x-cmd/deepseek/SKILL.zh.md b/data/x-cmd/deepseek/SKILL.zh.md deleted file mode 100644 index 0419720..0000000 --- a/data/x-cmd/deepseek/SKILL.zh.md +++ /dev/null @@ -1,49 +0,0 @@ ---- -name: deepseek -description: > - 集成 DeepSeek AI,用于高性能文本生成和专业推理任务。 - 核心场景:当用户想要使用 DeepSeek 的 V3 或推理模型进行编码、翻译或复杂逻辑处理时。 -license: MIT ---- - -# deepseek - DeepSeek AI 集成 - -`deepseek` 模块提供 DeepSeek AI 模型的命令行界面,包括强大的 V3 模型和专注于推理的模型。 - -## 激活时机 -- 当用户想要使用 DeepSeek 进行聊天、文本生成或编码任务时。 -- 当用户需要使用 `deepseek-reasoner` 模型进行复杂推理或数学解题时。 -- 当检查 DeepSeek 账户余额或管理 API 密钥时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 进行设置。 -- **模型选择**: 默认通常为 V3 (`@ds`)。对于需要深度思考的任务,请使用 `--model deepseek-reasoner`。 -- **余额监控**: 使用 `balance` 子命令跟踪信用额度。 - -## 补充场景 -- **快捷别名**: 使用 `@ds` 进行快速交互。 -- **项目配置**: 使用 `--cfg` 选项按项目设置默认模型。 - -## 实战示例 - -### 与 DeepSeek V3 聊天 -```bash -# 提出一般性问题 -@ds "解释量子纠缠的概念" -``` - -### 逻辑推理 -```bash -# 使用推理模型处理复杂逻辑 -@ds --model deepseek-reasoner "解决这个复杂的概率问题:..." -``` - -### 检查账户余额 -```bash -# 查看当前余额 -x deepseek balance -``` - -## 交付验证清单 -- [ ] 确保已初始化 DeepSeek API 密钥。 -- [ ] 确认推理模型是否更适合当前任务。 diff --git a/data/x-cmd/deno/SKILL.md b/data/x-cmd/deno/SKILL.md deleted file mode 100644 index 67c1060..0000000 --- a/data/x-cmd/deno/SKILL.md +++ /dev/null @@ -1,28 +0,0 @@ ---- -name: deno -description: > - Enhanced Deno module for running modern JavaScript and TypeScript with zero setup. - Core Scenario: When the user needs to execute TypeScript scripts or manage Deno-based applications. -license: MIT ---- - -# deno - Modern JS/TS Runtime - -The `deno` module provides a powerful interface for the Deno runtime, enabling secure and fast execution of JavaScript and TypeScript scripts without complex environment setup. - -## When to Activate -- When executing TypeScript or JavaScript files. -- When managing Deno dependencies or permissions for scripts. -- When needing a modern, secure runtime for one-off tasks. - -## Patterns & Examples - -### Run TypeScript -```bash -# Execute a local TypeScript file -x deno run script.ts -``` - -## Checklist -- [ ] Confirm if script permissions (network, file access) are required. -- [ ] Verify if TypeScript or JavaScript is being used. diff --git a/data/x-cmd/deno/SKILL.zh.md b/data/x-cmd/deno/SKILL.zh.md deleted file mode 100644 index f35f8be..0000000 --- a/data/x-cmd/deno/SKILL.zh.md +++ /dev/null @@ -1,28 +0,0 @@ ---- -name: deno -description: > - 增强型 Deno 模块,用于零设置运行现代 JavaScript 和 TypeScript。 - 核心场景:当用户需要执行 TypeScript 脚本或管理基于 Deno 的应用时。 -license: MIT ---- - -# deno - 现代 JS/TS 运行时 - -`deno` 模块为 Deno 运行时提供了一个强大的界面,能够安全、快速地执行 JavaScript 和 TypeScript 脚本,无需复杂的环境设置。 - -## 激活时机 -- 当执行 TypeScript 或 JavaScript 文件时。 -- 当管理 Deno 依赖项或脚本权限时。 -- 当为一次性任务需要一个现代、安全的运行时时。 - -## 实战示例 - -### 运行 TypeScript -```bash -# 执行本地 TypeScript 文件 -x deno run script.ts -``` - -## 交付验证清单 -- [ ] 确认是否需要脚本权限(网络、文件访问)。 -- [ ] 验证使用的是 TypeScript 还是 JavaScript。 diff --git a/data/x-cmd/dingtalk/SKILL.md b/data/x-cmd/dingtalk/SKILL.md deleted file mode 100644 index b75c7b4..0000000 --- a/data/x-cmd/dingtalk/SKILL.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: dingtalk -description: > - Send messages to DingTalk groups using bot webhooks, supporting Markdown and interactive cards. - Core Scenario: When the user needs to automate notifications or send interactive messages to a DingTalk group. -license: MIT ---- - -# dingtalk - DingTalk Bot Utility - -The `dingtalk` module allows users to send text, markdown, and rich text card messages to DingTalk groups via bot webhooks. - -## When to Activate -- When sending automated build or status notifications to a DingTalk group. -- When using Markdown for formatted alerts (titles, lists, bold text). -- When creating interactive action cards with links and images. - -## Core Principles & Rules -- **Webhook Required**: Set the bot `webhook` url via `init` or `--cfg`. -- **Markdown Support**: Use `--markdown` for styled messages. -- **Card Messages**: Use `--richtext` and `--card` for complex interactive layouts. - -## Patterns & Examples - -### Send Markdown -```bash -# Send a styled markdown notification to DingTalk -x dingtalk send --markdown --title "Release Note" "## Version v1.0.0\n- Bug fixes\n- New features" -``` - -### Send Action Card -```bash -# Send a rich text card with a link and an image -x dingtalk send --richtext --card "View Details" "https://x-cmd.com" "https://example.com/image.png" -``` - -## Checklist -- [ ] Ensure the DingTalk webhook URL is configured. -- [ ] Verify the message format (text/markdown/card) suits the content. diff --git a/data/x-cmd/dingtalk/SKILL.zh.md b/data/x-cmd/dingtalk/SKILL.zh.md deleted file mode 100644 index 6eba2ce..0000000 --- a/data/x-cmd/dingtalk/SKILL.zh.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: dingtalk -description: > - 使用机器人 Webhook 向钉钉群组发送消息,支持 Markdown 和交互式卡片。 - 核心场景:当用户需要向钉钉群组发送自动化通知或交互式消息时。 -license: MIT ---- - -# dingtalk - 钉钉机器人工具 - -`dingtalk` 模块允许用户通过机器人 Webhook 向钉钉群组发送文本、Markdown 和富文本卡片消息。 - -## 激活时机 -- 当向钉钉群组发送自动化的构建或状态通知时。 -- 当使用 Markdown 进行格式化警报(标题、列表、加粗文本)时。 -- 当创建带有链接和图片的交互式操作卡片时。 - -## 核心原则与规范 -- **需要 Webhook**: 通过 `init` 或 `--cfg` 设置机器人 `webhook` URL。 -- **Markdown 支持**: 使用 `--markdown` 发送带样式的消息。 -- **卡片消息**: 使用 `--richtext` 和 `--card` 实现复杂的交互式布局。 - -## 实战示例 - -### 发送 Markdown -```bash -# 向钉钉发送带样式的 Markdown 通知 -x dingtalk send --markdown --title "发布说明" "## 版本 v1.0.0\n- Bug 修复\n- 新功能" -``` - -### 发送操作卡片 -```bash -# 发送带有链接和图片的富文本卡片 -x dingtalk send --richtext --card "查看详情" "https://x-cmd.com" "https://example.com/image.png" -``` - -## 交付验证清单 -- [ ] 确保钉钉 Webhook URL 已配置。 -- [ ] 验证消息格式(文本/Markdown/卡片)是否适合内容。 diff --git a/data/x-cmd/discord/SKILL.md b/data/x-cmd/discord/SKILL.md deleted file mode 100644 index 5023eb9..0000000 --- a/data/x-cmd/discord/SKILL.md +++ /dev/null @@ -1,31 +0,0 @@ ---- -name: discord -description: > - Send messages to Discord channels using bot webhooks. - Core Scenario: When the user needs to automate notifications or send alerts to a Discord group. -license: MIT ---- - -# discord - Discord Bot Utility - -The `discord` module enables users to send text messages to Discord channels through bot webhooks, making it easy to integrate terminal alerts into Discord workflows. - -## When to Activate -- When sending automated alerts or notifications from terminal scripts to Discord. -- When managing Discord bot webhooks for different channels. - -## Core Principles & Rules -- **Webhook Required**: Remind users to configure the `webhook` url via `init` or `--cfg`. -- **Simplicity**: Currently focused on straightforward text message delivery. - -## Patterns & Examples - -### Send Message -```bash -# Send a text notification to the configured Discord channel -x discord send "Server backup completed successfully" -``` - -## Checklist -- [ ] Ensure the Discord webhook URL is configured. -- [ ] Verify the message content is appropriate for the target channel. diff --git a/data/x-cmd/discord/SKILL.zh.md b/data/x-cmd/discord/SKILL.zh.md deleted file mode 100644 index f150d5d..0000000 --- a/data/x-cmd/discord/SKILL.zh.md +++ /dev/null @@ -1,31 +0,0 @@ ---- -name: discord -description: > - 使用机器人 Webhook 向 Discord 频道发送消息。 - 核心场景:当用户需要自动化通知或向 Discord 群组发送警报时。 -license: MIT ---- - -# discord - Discord 机器人工具 - -`discord` 模块使用户能够通过机器人 Webhook 向 Discord 频道发送文本消息,从而轻松地将终端警报集成到 Discord 工作流中。 - -## 激活时机 -- 当从终端脚本向 Discord 发送自动化警报或通知时。 -- 当为不同频道管理 Discord 机器人 Webhook 时。 - -## 核心原则与规范 -- **需要 Webhook**: 提醒用户通过 `init` 或 `--cfg` 配置 `webhook` URL。 -- **简洁性**: 目前专注于直接的文本消息传递。 - -## 实战示例 - -### 发送消息 -```bash -# 向配置好的 Discord 频道发送文本通知 -x discord send "服务器备份已成功完成" -``` - -## 交付验证清单 -- [ ] 确保 Discord Webhook URL 已配置。 -- [ ] 验证消息内容是否适合目标频道。 diff --git a/data/x-cmd/dns/SKILL.md b/data/x-cmd/dns/SKILL.md deleted file mode 100644 index c64b8b1..0000000 --- a/data/x-cmd/dns/SKILL.md +++ /dev/null @@ -1,44 +0,0 @@ ---- -name: dns -description: > - Manage system DNS configurations, view current settings, and refresh DNS cache. - Core Scenario: When the user needs to audit DNS servers or resolve name resolution issues by flushing cache. -license: MIT ---- - -# dns - DNS Configuration & Management - -The `dns` module provides a simple interface for viewing current DNS settings, listing available DNS servers, and performing maintenance tasks like refreshing the system's DNS cache. - -## When to Activate -- When checking which DNS servers the system is currently using. -- When troubleshooting domain name resolution issues. -- When needing to flush/refresh the local DNS cache to pick up changes. - -## Core Principles & Rules -- **Visibility**: Consolidates DNS info into a single command (`current`). -- **Maintenance**: Use `refresh` to clear cached name resolution data. - -## Patterns & Examples - -### View Configuration -```bash -# Show the current system DNS settings -x dns current -``` - -### Refresh Cache -```bash -# Flush and refresh the system DNS cache -x dns refresh -``` - -### List Available Servers -```bash -# List all configured DNS server addresses -x dns ls -``` - -## Checklist -- [ ] Confirm if the user is investigating a resolution problem or just checking settings. -- [ ] Verify if a DNS cache flush is appropriate for the current issue. diff --git a/data/x-cmd/dns/SKILL.zh.md b/data/x-cmd/dns/SKILL.zh.md deleted file mode 100644 index 2a95b0e..0000000 --- a/data/x-cmd/dns/SKILL.zh.md +++ /dev/null @@ -1,44 +0,0 @@ ---- -name: dns -description: > - 管理系统 DNS 配置,查看当前设置并刷新 DNS 缓存。 - 核心场景:当用户需要审核 DNS 服务器或通过清理缓存解决名称解析问题时。 -license: MIT ---- - -# dns - DNS 配置与管理 - -`dns` 模块提供了一个简单的界面,用于查看当前 DNS 设置、列出可用的 DNS 服务器,并执行维护任务(如刷新系统的 DNS 缓存)。 - -## 激活时机 -- 当检查系统当前正在使用哪些 DNS 服务器时。 -- 当排除域名解析故障时。 -- 当需要刷新本地 DNS 缓存以应用更改时。 - -## 核心原则与规范 -- **可见性**: 将 DNS 信息合并到单个命令 (`current`) 中。 -- **维护**: 使用 `refresh` 清除缓存的名称解析数据。 - -## 实战示例 - -### 查看配置 -```bash -# 显示当前系统 DNS 设置 -x dns current -``` - -### 刷新缓存 -```bash -# 刷新系统 DNS 缓存 -x dns refresh -``` - -### 列出可用服务器 -```bash -# 列出所有已配置的 DNS 服务器地址 -x dns ls -``` - -## 交付验证清单 -- [ ] 确认用户是在调查解析问题还是仅仅检查设置。 -- [ ] 验证刷新 DNS 缓存是否适合当前问题。 diff --git a/data/x-cmd/docker/SKILL.md b/data/x-cmd/docker/SKILL.md deleted file mode 100644 index 839f298..0000000 --- a/data/x-cmd/docker/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: docker -description: > - Enhanced Docker management tool with support for mirror acceleration, structured data exports, and container analysis. - Core Scenario: When the user needs to manage Docker containers and images with optimized performance or export states. -license: MIT ---- - -# docker - Enhanced Docker Management - -The `docker` module provides an enhanced set of tools for Docker, offering mirror acceleration for faster image pulls, structured data exports (JSON/CSV), and interactive TUI explorers. - -## When to Activate -- When pulling images from Docker Hub or other registries (automatically uses mirrors). -- When exporting container, image, or volume lists to JSON/CSV for scripting. -- When performing interactive analysis of running containers using the `app` or `fz` modes. -- When managing Docker configurations and mirror registry settings. - -## Core Principles & Rules -- **Acceleration**: Transparently applies mirrors to speed up image pulls in specific regions. -- **Structured Data**: Prioritize `--json` or `--csv` for any subcommand when results are for automation. -- **TUI Mode**: Use `ps --app` or `fz` for a visual Docker dashboard experience. - -## Patterns & Examples - -### Interactive Dashboard -```bash -# Open an interactive TUI to manage running containers -x docker ps --app -``` - -### Mirror Configuration -```bash -# Set up Docker to use an optimized mirror registry -x docker mirror use ustc -``` - -### Export Image List -```bash -# Get all local images as a JSON array -x docker images --json -``` - -## Checklist -- [ ] Confirm if mirror acceleration is needed for image tasks. -- [ ] Verify the desired output format (Human-readable, JSON, CSV). -- [ ] Ensure Docker daemon is running on the host. diff --git a/data/x-cmd/docker/SKILL.zh.md b/data/x-cmd/docker/SKILL.zh.md deleted file mode 100644 index 90dcaf2..0000000 --- a/data/x-cmd/docker/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: docker -description: > - 增强型 Docker 管理工具,支持镜像加速、结构化数据导出和容器分析。 - 核心场景:当用户需要以优化后的性能管理 Docker 容器和镜像,或导出状态时。 -license: MIT ---- - -# docker - 增强型 Docker 管理 - -`docker` 模块为 Docker 提供了一套增强工具,包括用于加速镜像拉取的镜像加速、结构化数据导出(JSON/CSV)以及交互式 TUI 浏览器。 - -## 激活时机 -- 当从 Docker Hub 或其他注册表拉取镜像时(自动使用镜像加速)。 -- 当为脚本编写将容器、镜像或卷列表导出为 JSON/CSV 时。 -- 当使用 `app` 或 `fz` 模式对运行中的容器执行交互式分析时。 -- 当管理 Docker 配置和镜像加速设置时。 - -## 核心原则与规范 -- **加速**: 透明地应用镜像源,以在特定地区加速镜像拉取。 -- **结构化数据**: 当结果用于自动化时,为任何子命令优先使用 `--json` 或 `--csv`。 -- **TUI 模式**: 使用 `ps --app` 或 `fz` 获得视觉化的 Docker 仪表板体验。 - -## 实战示例 - -### 交互式仪表板 -```bash -# 打开交互式 TUI 来管理运行中的容器 -x docker ps --app -``` - -### 镜像加速配置 -```bash -# 设置 Docker 使用优化的镜像源 -x docker mirror use ustc -``` - -### 导出镜像列表 -```bash -# 以 JSON 数组形式获取所有本地镜像 -x docker images --json -``` - -## 交付验证清单 -- [ ] 确认镜像任务是否需要镜像加速。 -- [ ] 验证所需的输出格式(人类可读、JSON, CSV)。 -- [ ] 确保宿主机上的 Docker 守护进程正在运行。 diff --git a/data/x-cmd/doubao/SKILL.md b/data/x-cmd/doubao/SKILL.md deleted file mode 100644 index 1983138..0000000 --- a/data/x-cmd/doubao/SKILL.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: doubao -description: > - Integrate Doubao (ByteDance Ark) AI models for chat and text processing. - Core Scenario: When the user wants to use Doubao models for chat, reasoning, or text generation via CLI. -license: MIT ---- - -# doubao - Doubao AI Integration - -The `doubao` module provides a CLI interface for ByteDance's Doubao models through the Volcengine Ark platform. - -## When to Activate -- When the user wants to chat with Doubao models. -- When using specialized models like `doubao-seed`. -- When managing Doubao API keys and configurations. - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` for setup. -- **Model Selection**: Use the `--model` flag to specify a specific Doubao model version. -- **Quick Access**: Use the `@doubao` alias for rapid interaction. - -## Patterns & Examples - -### Chat with Doubao -```bash -# Ask a question to the default Doubao model -@doubao "What are the benefits of using x-cmd?" -``` - -### Use Specific Model -```bash -# Use a specific seed model version -@doubao --model doubao-seed-1-8-251228 "Explain this code: ..." -``` - -## Checklist -- [ ] Ensure the Doubao API key is configured. -- [ ] Verify the chosen model name is correct. diff --git a/data/x-cmd/doubao/SKILL.zh.md b/data/x-cmd/doubao/SKILL.zh.md deleted file mode 100644 index cda7e61..0000000 --- a/data/x-cmd/doubao/SKILL.zh.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: doubao -description: > - 集成豆包 (字节跳动火山方舟) AI 模型,用于聊天和文本处理。 - 核心场景:当用户想要通过 CLI 使用豆包模型进行聊天、推理或文本生成时。 -license: MIT ---- - -# doubao - 豆包 AI 集成 - -`doubao` 模块通过火山方舟平台为字节跳动的豆包模型提供命令行界面。 - -## 激活时机 -- 当用户想要与豆包模型聊天时。 -- 当使用 `doubao-seed` 等专业模型时。 -- 当管理豆包 API 密钥和配置时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 进行设置。 -- **模型选择**: 使用 `--model` 标志指定特定的豆包模型版本。 -- **快速访问**: 使用 `@doubao` 别名进行快速交互。 - -## 实战示例 - -### 与豆包聊天 -```bash -# 向默认豆包模型提问 -@doubao "使用 x-cmd 有什么好处?" -``` - -### 使用特定模型 -```bash -# 使用特定的种子模型版本 -@doubao --model doubao-seed-1-8-251228 "解释这段代码:..." -``` - -## 交付验证清单 -- [ ] 确保已配置豆包 API 密钥。 -- [ ] 验证所选模型名称是否正确。 diff --git a/data/x-cmd/elv/SKILL.md b/data/x-cmd/elv/SKILL.md deleted file mode 100644 index 2589a04..0000000 --- a/data/x-cmd/elv/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: elv -description: > - Enhanced interface for Elvish shell, supporting x-cmd integration, official documentation search, and AI commands. - Core Scenario: When the user needs to integrate x-cmd with Elvish or search Elvish documentation via terminal. -license: MIT ---- - -# elv - Elvish Shell Enhancement - -The `elv` module facilitates the use of the Elvish shell by providing seamless integration with x-cmd and quick lookup of Elvish documentation. - -## When to Activate -- When the user wants to use x-cmd tools (x, c, @gpt) within the Elvish shell. -- When searching elv.sh for syntax, modules, or alias definitions. -- When launching Elvish with zero-setup via the x-cmd package manager. - -## Core Principles & Rules -- **Integration**: Use `--setup` to add x-cmd to the Elvish environment. -- **Dynamic Acquisition**: Automatically fetches Elvish if it's not pre-installed. - -## Patterns & Examples - -### Inject x-cmd -```bash -# Set up x-cmd tools in the Elvish shell environment -x elv --setup -``` - -### Documentation Lookup -```bash -# Search for 'alias' examples on the Elvish website -x elv : alias -``` - -## Checklist -- [ ] Ensure the user is familiar with Elvish's structured data approach. -- [ ] Confirm if the Elvish configuration file needs permanent modification. diff --git a/data/x-cmd/elv/SKILL.zh.md b/data/x-cmd/elv/SKILL.zh.md deleted file mode 100644 index ccea380..0000000 --- a/data/x-cmd/elv/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: elv -description: > - Elvish Shell 的增强接口,支持 x-cmd 集成、官方文档搜索和 AI 命令生成。 - 核心场景:当用户需要将 x-cmd 与 Elvish 集成或通过终端搜索 Elvish 文档时。 -license: MIT ---- - -# elv - Elvish Shell 增强工具 - -`elv` 模块通过提供与 x-cmd 的无缝集成以及快速查找 Elvish 文档,简化了 Elvish Shell 的使用。 - -## 激活时机 -- 当用户想要在 Elvish Shell 中使用 x-cmd 工具(x, c, @gpt)时。 -- 当在 elv.sh 网站上搜索语法、模块或别名定义时。 -- 当通过 x-cmd 包管理器零设置启动 Elvish 时。 - -## 核心原则与规范 -- **集成**: 使用 `--setup` 将 x-cmd 添加到 Elvish 环境中。 -- **动态获取**: 如果未预装 Elvish,会自动获取。 - -## 实战示例 - -### 注入 x-cmd -```bash -# 在 Elvish Shell 环境中设置 x-cmd 工具 -x elv --setup -``` - -### 文档查找 -```bash -# 在 Elvish 网站上搜索 'alias' 示例 -x elv : alias -``` - -## 交付验证清单 -- [ ] 确保用户熟悉 Elvish 的结构化数据处理方式。 -- [ ] 确认是否需要对 Elvish 配置文件进行永久修改。 diff --git a/data/x-cmd/facl/SKILL.md b/data/x-cmd/facl/SKILL.md deleted file mode 100644 index 665dcf3..0000000 --- a/data/x-cmd/facl/SKILL.md +++ /dev/null @@ -1,33 +0,0 @@ ---- -name: facl -description: > - Retrieve and manage File Access Control Lists (ACL) in a more intuitive format. - Core Scenario: When the user needs to view or set fine-grained file permissions (ACLs) via terminal. -license: MIT ---- - -# facl - File Access Control List Management - -The `facl` module simplifies the interaction with File Access Control Lists (ACL), providing an easier way to view and set complex file permissions beyond standard Unix rwx bits. - -## When to Activate -- When the user needs to audit fine-grained permissions on files or directories. -- When setting specific user or group permissions using ACLs. - -## Core Principles & Rules -- **Intuitive Display**: Designed to show ACL entries in a clear, readable format. -- **Standard Operations**: Supports both `get` (view) and `set` (modify) subcommands. - -## Patterns & Examples - -### View ACL -```bash -# Get the ACL for the current directory -x facl -# Get the ACL for a specific file -x facl myfile.txt -``` - -## Checklist -- [ ] Confirm if the user intends to view or modify ACLs. -- [ ] Verify the target file or directory path. diff --git a/data/x-cmd/facl/SKILL.zh.md b/data/x-cmd/facl/SKILL.zh.md deleted file mode 100644 index db6e96c..0000000 --- a/data/x-cmd/facl/SKILL.zh.md +++ /dev/null @@ -1,33 +0,0 @@ ---- -name: facl -description: > - 以更直观的格式获取和管理文件访问控制列表 (ACL)。 - 核心场景:当用户需要通过终端查看或设置细粒度的文件权限 (ACL) 时。 -license: MIT ---- - -# facl - 文件访问控制列表管理 - -`facl` 模块简化了与文件访问控制列表 (ACL) 的交互,提供了一种更简单的方法来查看和设置超出标准 Unix rwx 位之外的复杂文件权限。 - -## 激活时机 -- 当用户需要审核文件或目录上的细粒度权限时。 -- 当使用 ACL 设置特定的用户或组权限时。 - -## 核心原则与规范 -- **直观显示**: 旨在以清晰、易读的格式显示 ACL 条目。 -- **标准操作**: 支持 `get`(查看)和 `set`(修改)子命令。 - -## 实战示例 - -### 查看 ACL -```bash -# 获取当前目录的 ACL -x facl -# 获取特定文件的 ACL -x facl myfile.txt -``` - -## 交付验证清单 -- [ ] 确认用户是打算查看还是修改 ACL。 -- [ ] 验证目标文件或目录路径。 diff --git a/data/x-cmd/feishu/SKILL.md b/data/x-cmd/feishu/SKILL.md deleted file mode 100644 index a68b2bf..0000000 --- a/data/x-cmd/feishu/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: feishu -description: > - Send messages to Feishu (Lark) groups using bot webhooks, supporting rich text and cards. - Core Scenario: When the user needs to send automated notifications or rich text messages to a Feishu group. -license: MIT ---- - -# feishu - Feishu (Lark) Bot Utility - -The `feishu` module enables users to send text and interactive rich text messages to Feishu groups via bot webhooks. - -## When to Activate -- When sending automated status updates to a Feishu group. -- When creating rich text notifications with clickable links and titles. -- When integrating Feishu alerts into terminal-based automation scripts. - -## Core Principles & Rules -- **Webhook Required**: Configure the `webhook` url via `init` or `--cfg`. -- **Rich Text Support**: Use `--richtext` to include titles, links, and structured text. - -## Patterns & Examples - -### Send Simple Text -```bash -# Send a basic text message to Feishu -x feishu send --text "New version released" -``` - -### Send Rich Text -```bash -# Send a message with a title and an embedded link -x feishu send --richtext --title "Project Update" --text "Check details here:" --url "website" "https://x-cmd.com" -``` - -## Checklist -- [ ] Verify the Feishu webhook URL is set up. -- [ ] Confirm if rich text formatting is required for the message. diff --git a/data/x-cmd/feishu/SKILL.zh.md b/data/x-cmd/feishu/SKILL.zh.md deleted file mode 100644 index d167869..0000000 --- a/data/x-cmd/feishu/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: feishu -description: > - 使用机器人 Webhook 向飞书 (Feishu/Lark) 群组发送消息,支持富文本和卡片。 - 核心场景:当用户需要向飞书群组发送自动化通知或富文本消息时。 -license: MIT ---- - -# feishu - 飞书机器人工具 - -`feishu` 模块使用户能够通过机器人 Webhook 向飞书群组发送文本和交互式富文本消息。 - -## 激活时机 -- 当向飞书群组发送自动化状态更新时。 -- 当创建带有可点击链接和标题的富文本通知时。 -- 当将飞书警报集成到基于终端的自动化脚本中时。 - -## 核心原则与规范 -- **需要 Webhook**: 通过 `init` 或 `--cfg` 配置 `webhook` URL。 -- **富文本支持**: 使用 `--richtext` 包含标题、链接和结构化文本。 - -## 实战示例 - -### 发送简单文本 -```bash -# 向飞书发送基础文本消息 -x feishu send --text "新版本已发布" -``` - -### 发送富文本 -```bash -# 发送带有标题和嵌入链接的消息 -x feishu send --richtext --title "项目更新" --text "在此处查看详情:" --url "官网" "https://x-cmd.com" -``` - -## 交付验证清单 -- [ ] 验证飞书 Webhook URL 已设置。 -- [ ] 确认消息是否需要富文本格式。 diff --git a/data/x-cmd/ffmpeg/SKILL.md b/data/x-cmd/ffmpeg/SKILL.md deleted file mode 100644 index 881195a..0000000 --- a/data/x-cmd/ffmpeg/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: ffmpeg -description: > - Powerful multimedia processing tool for converting, recording, and streaming audio and video. - Core Scenario: When the user needs to convert media formats, extract audio, or perform complex video editing via CLI. -license: MIT ---- - -# ffmpeg - Comprehensive Multimedia Processor - -The `ffmpeg` module provides a powerful CLI for processing audio, video, and other multimedia files. It ensures availability by automatically installing ffmpeg via pixi if it is missing from the system. - -## When to Activate -- When converting between different video or audio formats (e.g., MP4 to AVI). -- When extracting audio tracks from video files (e.g., MP4 to MP3). -- When performing complex media operations like resizing, re-encoding, or applying filters. -- When needing to stream multimedia content from the terminal. - -## Core Principles & Rules -- **Zero-Setup**: Automatically handles installation via `pixi` if necessary. -- **Pass-through**: Supports all standard `ffmpeg` arguments via the `--` or `cmd` subcommand. -- **Input/Output**: Always clarify the `-i` (input) and target output paths. - -## Patterns & Examples - -### Format Conversion -```bash -# Convert a video from mp4 to avi -x ffmpeg -i input.mp4 output.avi -``` - -### Extract Audio -```bash -# Extract audio from a video file and save as MP3 -x ffmpeg -i input.mp4 -vn output.mp3 -``` - -### Direct Command -```bash -# Execute a raw ffmpeg command string -x ffmpeg --cmd -version -``` - -## Checklist -- [ ] Confirm the input file exists and is accessible. -- [ ] Verify the desired output format and codec settings. -- [ ] Ensure appropriate disk space for large media operations. diff --git a/data/x-cmd/ffmpeg/SKILL.zh.md b/data/x-cmd/ffmpeg/SKILL.zh.md deleted file mode 100644 index c50cc26..0000000 --- a/data/x-cmd/ffmpeg/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: ffmpeg -description: > - 强大的多媒体处理工具,用于转换、录制和流式传输音频和视频。 - 核心场景:当用户需要转换媒体格式、提取音频或通过 CLI 执行复杂的视频编辑时。 -license: MIT ---- - -# ffmpeg - 全能多媒体处理器 - -`ffmpeg` 模块提供了一个强大的命令行界面,用于处理音频、视频和其他多媒体文件。它通过在系统缺失 ffmpeg 时利用 pixi 自动安装,确保了工具的可用性。 - -## 激活时机 -- 当在不同的视频或音频格式之间进行转换时(例如 MP4 转 AVI)。 -- 当从视频文件中提取音频轨道时(例如 MP4 转 MP3)。 -- 当执行复杂的媒体操作(如调整大小、重新编码或应用过滤器)时。 -- 当需要从终端流式传输多媒体内容时。 - -## 核心原则与规范 -- **零设置**: 如有必要,会自动通过 `pixi` 处理安装。 -- **透明透传**: 通过 `--` 或 `cmd` 子命令支持所有标准的 `ffmpeg` 参数。 -- **输入/输出**: 始终明确 `-i`(输入)和目标输出路径。 - -## 实战示例 - -### 格式转换 -```bash -# 将视频从 mp4 转换为 avi 格式 -x ffmpeg -i input.mp4 output.avi -``` - -### 提取音频 -```bash -# 从视频文件中提取音频并保存为 MP3 -x ffmpeg -i input.mp4 -vn output.mp3 -``` - -### 直接执行 -```bash -# 执行原始的 ffmpeg 命令字符串 -x ffmpeg --cmd -version -``` - -## 交付验证清单 -- [ ] 确认输入文件存在且可访问。 -- [ ] 验证所需的输出格式和编解码器设置。 -- [ ] 确保有足够的磁盘空间进行大型媒体操作。 diff --git a/data/x-cmd/find/SKILL.md b/data/x-cmd/find/SKILL.md deleted file mode 100644 index 3119673..0000000 --- a/data/x-cmd/find/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: find -description: > - Search for files and directories in a directory hierarchy with an interactive FZF application. - Core Scenario: When the user needs to find files based on name, size, type, or other metadata. -license: MIT ---- - -# find - Enhanced File Search - -The `find` module extends the standard system find utility, offering an interactive FZF-powered application for rapid file discovery and metadata-based filtering. - -## When to Activate -- When the user wants to locate files or directories based on patterns or attributes. -- When an interactive, searchable list of files is required (`--fzfapp`). -- When finding files by size (e.g., larger than 100MB) or modification time. - -## Core Principles & Rules -- **Interactive Mode**: Automatically launches a TUI app for easy browsing if no specific criteria are given. -- **Attribute-Based**: Supports standard find flags like `-type`, `-name`, `-size`, and `-mtime`. -- **Actionable Results**: Can execute commands on found items using `-exec`. - -## Patterns & Examples - -### Interactive Discovery -```bash -# Open interactive FZF app to find files or folders -x find -``` - -### Search by Name -```bash -# Find all text files in the current directory -x find . -name "*.txt" -``` - -### Large File Search -```bash -# Find files larger than 100MB in the home directory -x find /home -size +100M -``` - -## Checklist -- [ ] Confirm the search path and criteria (name, size, type). -- [ ] Verify if an interactive view or direct command execution (`-exec`) is needed. -- [ ] Ensure the user is aware of the recursive nature of the search. diff --git a/data/x-cmd/find/SKILL.zh.md b/data/x-cmd/find/SKILL.zh.md deleted file mode 100644 index 7dbc22f..0000000 --- a/data/x-cmd/find/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: find -description: > - 在目录层级中搜索文件和目录,具有交互式 FZF 应用程序。 - 核心场景:当用户需要根据名称、大小、类型或其他元数据查找文件时。 -license: MIT ---- - -# find - 增强型文件查找 - -`find` 模块扩展了标准的系统 find 实用程序,提供了一个由 FZF 驱动的交互式应用,用于快速发现文件和基于元数据的过滤。 - -## 激活时机 -- 当用户想要根据模式或属性定位文件或目录时。 -- 当需要交互式的、可搜索的文件列表时 (`--fzfapp`)。 -- 当按大小(如大于 100MB)或修改时间查找文件时。 - -## 核心原则与规范 -- **交互模式**: 如果未给出特定标准,会自动启动 TUI 应用以便于浏览。 -- **基于属性**: 支持标准的 find 标志,如 `-type`, `-name`, `-size` 和 `-mtime`。 -- **可操作结果**: 可以使用 `-exec` 对找到的项目执行命令。 - -## 实战示例 - -### 交互式发现 -```bash -# 打开交互式 FZF 应用以查找文件或文件夹 -x find -``` - -### 按名称搜索 -```bash -# 查找当前目录下的所有文本文件 -x find . -name "*.txt" -``` - -### 大文件搜索 -```bash -# 在家目录中查找大于 100MB 的文件 -x find /home -size +100M -``` - -## 交付验证清单 -- [ ] 确认搜索路径和标准(名称、大小、类型)。 -- [ ] 验证需要的是交互式视图还是直接的命令执行 (`-exec`)。 -- [ ] 确保用户知晓搜索的递归性质。 diff --git a/data/x-cmd/fish/SKILL.md b/data/x-cmd/fish/SKILL.md deleted file mode 100644 index 9024218..0000000 --- a/data/x-cmd/fish/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: fish -description: > - Enhanced interface for Fish shell, offering x-cmd integration, documentation search, and AI assistant features. - Core Scenario: When the user needs to set up x-cmd in Fish or search the official Fish documentation via CLI. -license: MIT ---- - -# fish - Friendly Interactive Shell Enhancement - -The `fish` module optimizes the experience for Fish shell users by simplifying x-cmd integration and providing rapid access to the official Fish documentation. - -## When to Activate -- When the user wants to use x-cmd's full suite of tools (x, c, @gpt) within Fish. -- When searching fishshell.com for specific commands or configuration tips. -- When managing Fish-specific environment settings and aliases. - -## Core Principles & Rules -- **Configuration**: Use `setup` to modify `config.fish` for permanent x-cmd integration. -- **Search Support**: Leverage the `:` prefix for high-speed documentation retrieval. - -## Patterns & Examples - -### Install x-cmd in Fish -```bash -# Add x-cmd environment variables and functions to Fish -x fish setup -``` - -### Search Fish Docs -```bash -# Interactively search fishshell.com for alias information -x fish : alias -``` - -## Checklist -- [ ] Confirm if the user wants to modify their Fish config permanently. -- [ ] Verify if Fish is already the active shell. diff --git a/data/x-cmd/fish/SKILL.zh.md b/data/x-cmd/fish/SKILL.zh.md deleted file mode 100644 index 21d6866..0000000 --- a/data/x-cmd/fish/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: fish -description: > - Fish Shell 的增强接口,提供 x-cmd 集成、文档搜索和 AI 助手功能。 - 核心场景:当用户需要在 Fish 中设置 x-cmd 或通过 CLI 搜索官方 Fish 文档时。 -license: MIT ---- - -# fish - 友好交互式 Shell 增强工具 - -`fish` 模块通过简化 x-cmd 集成并提供对官方 Fish 文档的快速访问,为 Fish Shell 用户优化了体验。 - -## 激活时机 -- 当用户想要在 Fish 中使用 x-cmd 的全套工具(x, c, @gpt)时。 -- 当在 fishshell.com 网站上搜索特定命令或配置提示时。 -- 当管理 Fish 特定的环境设置和别名时。 - -## 核心原则与规范 -- **配置**: 使用 `setup` 修改 `config.fish` 以实现永久的 x-cmd 集成。 -- **搜索支持**: 利用 `:` 前缀进行高速文档检索。 - -## 实战示例 - -### 在 Fish 中安装 x-cmd -```bash -# 将 x-cmd 环境变量和函数添加到 Fish 中 -x fish setup -``` - -### 搜索 Fish 文档 -```bash -# 交互式搜索 fishshell.com 关于别名的信息 -x fish : alias -``` - -## 交付验证清单 -- [ ] 确认用户是否想要永久修改其 Fish 配置。 -- [ ] 验证 Fish 是否已经是当前激活的 Shell。 diff --git a/data/x-cmd/fjo/SKILL.md b/data/x-cmd/fjo/SKILL.md deleted file mode 100644 index 9d18bf4..0000000 --- a/data/x-cmd/fjo/SKILL.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: fjo -description: > - CLI interface for Forgejo, the community-driven self-hosted software development platform. - Core Scenario: When the user needs to manage repositories or issues on a Forgejo instance. -license: MIT ---- - -# fjo - Forgejo CLI Browser - -The `fjo` module provides a dedicated CLI for Forgejo instances, supporting repository lifecycle management and collaboration features. - -## When to Activate -- When working with Forgejo-based git hosting services. -- When managing issues or pull requests on Forgejo instances. - -## Patterns & Examples - -### List Issues -```bash -# View open issues for a specific Forgejo repository -x fjo issue ls owner/repo -``` - -## Checklist -- [ ] Confirm the Forgejo instance URL and credentials. diff --git a/data/x-cmd/fjo/SKILL.zh.md b/data/x-cmd/fjo/SKILL.zh.md deleted file mode 100644 index 7fd4cb1..0000000 --- a/data/x-cmd/fjo/SKILL.zh.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: fjo -description: > - Forgejo 的命令行界面,这是一个社区驱动的自托管软件开发平台。 - 核心场景:当用户需要在 Forgejo 实例上管理仓库或 Issue 时。 -license: MIT ---- - -# fjo - Forgejo 命令行浏览器 - -`fjo` 模块为 Forgejo 实例提供了专用的命令行界面,支持仓库生命周期管理和协作功能。 - -## 激活时机 -- 当使用基于 Forgejo 的 Git 托管服务时。 -- 当在 Forgejo 实例上管理 Issue 或 Pull Request 时。 - -## 实战示例 - -### 列出 Issue -```bash -# 查看特定 Forgejo 仓库的开启状态 Issue -x fjo issue ls owner/repo -``` - -## 交付验证清单 -- [ ] 确认 Forgejo 实例 URL 和凭据。 diff --git a/data/x-cmd/free/SKILL.md b/data/x-cmd/free/SKILL.md deleted file mode 100644 index ccf00f9..0000000 --- a/data/x-cmd/free/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: free -description: > - Display memory usage with support for CSV/TSV formats and cross-platform (Linux/macOS) compatibility. - Core Scenario: When the user needs to monitor RAM usage, buffer/cache, or export memory data for automation. -license: MIT ---- - -# free - Memory Usage Reporting - -The `free` module enhances the standard memory reporting tool by adding structured data support (CSV/TSV) and providing a native implementation for macOS users who typically lack the `free` command. - -## When to Activate -- When monitoring total, used, and free system memory. -- When performing memory usage analysis in scripts using CSV/TSV formats. -- When checking memory compression details on macOS. -- When needing periodic memory reporting (polling). - -## Core Principles & Rules -- **Cross-Platform**: Provides a consistent interface for memory stats on both Linux and macOS. -- **Automation-Ready**: Supports `--csv` and `--tsv` for easy parsing without headers. -- **Real-Time Polling**: Use `-s` and `-c` to repeat the output at intervals. - -## Patterns & Examples - -### Human-Readable View -```bash -# Display system memory usage in a clear format -x free -``` - -### Export for Scripting -```bash -# Get memory data as a single-line CSV without headers -x free --csv --no-header -``` - -### Periodic Monitoring -```bash -# Refresh memory stats every 2 seconds for 10 iterations -x free -s 2 -c 10 -``` - -## Checklist -- [ ] Verify if the output needs to be structured (CSV/TSV) for a script. -- [ ] Confirm if headers should be included or omitted. -- [ ] Ensure the correct polling interval is set if monitoring over time. diff --git a/data/x-cmd/free/SKILL.zh.md b/data/x-cmd/free/SKILL.zh.md deleted file mode 100644 index 9bc55be..0000000 --- a/data/x-cmd/free/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: free -description: > - 显示内存使用情况,支持 CSV/TSV 格式和跨平台 (Linux/macOS) 兼容性。 - 核心场景:当用户需要监控 RAM 使用、缓冲区/缓存,或为自动化导出内存数据时。 -license: MIT ---- - -# free - 内存使用报告 - -`free` 模块增强了标准的内存报告工具,增加了结构化数据支持 (CSV/TSV),并为通常缺少 `free` 命令的 macOS 用户提供了原生实现。 - -## 激活时机 -- 当监控系统的总内存、已用内存和空闲内存时。 -- 当在脚本中使用 CSV/TSV 格式执行内存使用分析时。 -- 当在 macOS 上检查内存压缩详情时。 -- 当需要定期内存报告(轮询)时。 - -## 核心原则与规范 -- **跨平台**: 为 Linux 和 macOS 上的内存统计提供一致的界面。 -- **自动化就绪**: 支持 `--csv` 和 `--tsv`,以便在没有表头的情况下轻松解析。 -- **实时轮询**: 使用 `-s` 和 `-c` 以固定间隔重复输出。 - -## 实战示例 - -### 人类可读视图 -```bash -# 以清晰的格式显示系统内存使用情况 -x free -``` - -### 脚本导出 -```bash -# 获取不带表头的单行 CSV 格式内存数据 -x free --csv --no-header -``` - -### 定期监控 -```bash -# 每 2 秒刷新一次内存统计信息,共执行 10 次 -x free -s 2 -c 10 -``` - -## 交付验证清单 -- [ ] 验证输出是否需要为脚本提供结构化格式 (CSV/TSV)。 -- [ ] 确认是否应包含或省略表头。 -- [ ] 如果是进行持续监控,请确保设置了正确的轮询间隔。 diff --git a/data/x-cmd/fsiter/SKILL.md b/data/x-cmd/fsiter/SKILL.md deleted file mode 100644 index e9c3112..0000000 --- a/data/x-cmd/fsiter/SKILL.md +++ /dev/null @@ -1,57 +0,0 @@ ---- -name: fsiter -description: > - Flexible file system iteration and search tool with DFS support and filtering. - Core Scenario: When the user needs to traverse directory trees, filter files/folders, or check for existence/emptiness. -license: MIT ---- - -# fsiter - File System Iteration & Search - -The `fsiter` module is a powerful alternative to `ls` and `find`, providing flexible filtering, depth-first search (DFS) traversal, and script-friendly queries. - -## When to Activate -- When the user wants to list specific file types (e.g., hidden-only, files-only). -- When performing recursive directory traversal with depth limits. -- When checking if a directory is empty or if specific patterns exist. -- When counting files or directories in a tree for reporting. - -## Core Principles & Rules -- **Depth-First Search**: Use `--dfs` for controlled tree traversal with custom callbacks. -- **Efficient Filtering**: Combine `--file`, `--folder`, and `--hidden` to pinpoint targets. -- **Script-Ready**: Subcommands like `--filecount` and `--dirempty` provide reliable exit codes and values for automation. - -## Additional Scenarios -- **Existence Check**: Quickly verify if patterns exist using `--exist`. -- **Recursive Processing**: Use `--dfs` with depth limits to process files in specific tree levels. - -## Patterns & Examples - -### List Specific Items -```bash -# List all hidden files in the current directory -x fsiter --ls --file --hidden-only -``` - -### Depth-Limited Search -```bash -# Traverse the tree up to 2 levels deep and echo item names -x fsiter --dfs /path/to/dir 0 2 echo -``` - -### Query Statistics -```bash -# Count items in a log directory -x fsiter --filecount /var/log -``` - -### Conditional Logic -```bash -# Delete a directory only if it is empty -x fsiter --dirempty ./temp && rmdir ./temp -``` - -## Checklist -- [ ] Confirm if the search should include hidden items. -- [ ] Verify the required depth for DFS traversal. -- [ ] Ensure the callback command for `--dfs` is correctly specified. diff --git a/data/x-cmd/fsiter/SKILL.zh.md b/data/x-cmd/fsiter/SKILL.zh.md deleted file mode 100644 index dd8e9a2..0000000 --- a/data/x-cmd/fsiter/SKILL.zh.md +++ /dev/null @@ -1,57 +0,0 @@ ---- -name: fsiter -description: > - 灵活的文件系统迭代与搜索工具,支持 DFS(深度优先搜索)和过滤。 - 核心场景:当用户需要遍历目录树、过滤文件/文件夹,或检查是否存在/是否为空时。 -license: MIT ---- - -# fsiter - 文件系统迭代与搜索 - -`fsiter` 模块是 `ls` 和 `find` 的强大替代方案,提供了灵活的过滤、深度优先搜索 (DFS) 遍历以及对脚本友好的查询功能。 - -## 激活时机 -- 当用户想要列出特定的文件类型(例如,仅隐藏文件、仅文件)时。 -- 当执行带有深度限制的递归目录遍历时。 -- 当检查目录是否为空或特定模式是否存在时。 -- 当统计目录树中的文件或目录数量以生成报告时。 - -## 核心原则与规范 -- **深度优先搜索**: 使用 `--dfs` 进行受控的树遍历,并带有自定义回调。 -- **高效过滤**: 结合 `--file`、`--folder` 和 `--hidden` 来精确定位目标。 -- **脚本就绪**: `--filecount` 和 `--dirempty` 等子命令为自动化提供了可靠的退出码和数值。 - -## 补充场景 -- **存在性检查**: 使用 `--exist` 快速验证模式是否存在。 -- **递归处理**: 使用带有深度限制的 `--dfs` 处理特定树层级的文件。 - -## 实战示例 - -### 列出特定项目 -```bash -# 列出当前目录下的所有隐藏文件 -x fsiter --ls --file --hidden-only -``` - -### 深度限制搜索 -```bash -# 遍历目录树,深度限制为 2 层,并回显项目名称 -x fsiter --dfs /path/to/dir 0 2 echo -``` - -### 查询统计 -```bash -# 统计日志目录中的项目数量 -x fsiter --filecount /var/log -``` - -### 条件逻辑 -```bash -# 仅当目录为空时才删除目录 -x fsiter --dirempty ./temp && rmdir ./temp -``` - -## 交付验证清单 -- [ ] 确认搜索是否应包含隐藏项。 -- [ ] 验证 DFS 遍历所需的深度。 -- [ ] 确保正确指定了 `--dfs` 的回调命令。 diff --git a/data/x-cmd/gcal/SKILL.md b/data/x-cmd/gcal/SKILL.md deleted file mode 100644 index e5997cd..0000000 --- a/data/x-cmd/gcal/SKILL.md +++ /dev/null @@ -1,44 +0,0 @@ ---- -name: gcal -description: > - Enhanced interface for GNU gcal, supporting holiday configurations and custom calendar formats. - Core Scenario: When the user needs a highly configurable calendar with national holiday support. -license: MIT ---- - -# gcal - GNU gcal Enhancement - -The `gcal` module provides a powerful wrapper around GNU gcal, making it easy to display calendars with localized holiday support and customizable formats. - -## When to Activate -- When the user wants to print monthly or yearly calendars. -- When needing holiday information for specific countries (e.g., China). -- When initializing or managing calendar preference settings. - -## Core Principles & Rules -- **Holiday Awareness**: Use the `cc` subcommand to list and select national holiday codes. -- **Persistent Config**: Use `cfg` to maintain settings like country codes across sessions. - -## Patterns & Examples - -### Print Current Month -```bash -# Display the calendar for the current month -x gcal -``` - -### Yearly Calendar -```bash -# Display the full calendar for 2025 -x gcal 2025 -``` - -### Configure Holidays -```bash -# Set gcal to use Chinese public holidays -x gcal --cfg cc=CN -``` - -## Checklist -- [ ] Confirm if a monthly or yearly view is needed. -- [ ] Verify if a specific country's holidays should be included. diff --git a/data/x-cmd/gcal/SKILL.zh.md b/data/x-cmd/gcal/SKILL.zh.md deleted file mode 100644 index 266aaa5..0000000 --- a/data/x-cmd/gcal/SKILL.zh.md +++ /dev/null @@ -1,44 +0,0 @@ ---- -name: gcal -description: > - GNU gcal 的增强型接口,支持节假日配置和自定义日历格式。 - 核心场景:当用户需要一个具有国家节假日支持的高度可配置日历时。 -license: MIT ---- - -# gcal - GNU gcal 增强工具 - -`gcal` 模块为 GNU gcal 提供了一个强大的包装器,可以轻松显示具有本地化节假日支持和可自定义格式的日历。 - -## 激活时机 -- 当用户想要打印月历或年历时。 -- 当需要特定国家(如中国)的节假日信息时。 -- 当初始化或管理日历首选项设置时。 - -## 核心原则与规范 -- **节假日感知**: 使用 `cc` 子命令列出并选择国家节假日代码。 -- **持久化配置**: 使用 `cfg` 在不同会话之间维护国家代码等设置。 - -## 实战示例 - -### 打印本月 -```bash -# 显示当前月份的日历 -x gcal -``` - -### 年历 -```bash -# 显示 2025 年的全年日历 -x gcal 2025 -``` - -### 配置节假日 -```bash -# 设置 gcal 以使用中国法定节假日 -x gcal --cfg cc=CN -``` - -## 交付验证清单 -- [ ] 确认需要的是月视图还是年视图。 -- [ ] 验证是否应包含特定国家的节假日。 diff --git a/data/x-cmd/gddy/SKILL.md b/data/x-cmd/gddy/SKILL.md deleted file mode 100644 index 8137d9b..0000000 --- a/data/x-cmd/gddy/SKILL.md +++ /dev/null @@ -1,55 +0,0 @@ ---- -name: gddy -description: > - GoDaddy domain management tool. - Core Scenario: When AI needs to check domain availability, list owned domains, or update DNS records. -license: MIT ---- - -# x gddy - GoDaddy Domain Management (AI Optimized) - -The `x gddy` module allows managing your GoDaddy domains via the command line. It's ideal for DNS maintenance and domain querying within AI workflows. - -## When to Activate -- When checking if a specific domain is available for purchase. -- When listing all domains owned by the GoDaddy account. -- When dynamically updating DNS resolution records (e.g., A records, CNAME). - -## Core Principles & Rules -- **Non-interactive First**: Use subcommands directly, avoiding unnecessary UI confirmations. -- **Environment Requirements**: Requires API Key and Secret. If not configured, AI should guide the user through initialization. -- **Configuration Guidance**: - - Direct the user to the GoDaddy Developer portal to get API info. - - Suggest the user run `x gddy init` for configuration. - -## Patterns & Examples - -### Check Domain Availability -```bash -# Check if example.com is available for purchase -x gddy search example.com -``` - -### List Owned Domains -```bash -# List all domains under the account (non-interactive) -x gddy domain ls -``` - -### Manage DNS Records -```bash -# View detailed info for a specific domain -x gddy domain info example.com - -# Add an A record for a domain -x gddy domain record add --name "www" --data "1.2.3.4" example.com -``` - -## Configuration Guide (for AI) -If an API error occurs or configuration is missing, provide this guidance to the user: -> Please obtain your API Key and Secret from the GoDaddy Developer portal (https://developer.godaddy.com/keys), then run the following command in your terminal to initialize: -> `x gddy init` - -## Checklist -- [ ] Confirm if API credentials are valid. -- [ ] Confirm before modifying DNS records or performing purchases. diff --git a/data/x-cmd/gddy/SKILL.zh.md b/data/x-cmd/gddy/SKILL.zh.md deleted file mode 100644 index 5dd8a7f..0000000 --- a/data/x-cmd/gddy/SKILL.zh.md +++ /dev/null @@ -1,55 +0,0 @@ ---- -name: gddy -description: > - GoDaddy 域名管理工具。 - 核心场景:当 AI 需要查询域名可用性、列出持有域名或更新 DNS 记录时。 -license: MIT ---- - -# x gddy - GoDaddy 域名管理 (AI 优化版) - -`x gddy` 模块允许通过命令行管理您的 GoDaddy 域名。它非常适合在脚本或 AI 工作流中执行 DNS 记录维护和域名查询。 - -## 激活时机 -- 当需要查询某个域名是否可以购买时。 -- 当需要列出当前 GoDaddy 账号下持有的所有域名时。 -- 当需要动态更新域名的 DNS 解析记录(如 A 记录、CNAME)时。 - -## 核心原则与规范 -- **非交互优先**: 直接使用子命令,避免不必要的 UI 确认。 -- **环境要求**: 需要 API Key 和 Secret。如果未配置,AI 应引导用户初始化。 -- **配置引导**: - - 引导用户访问 GoDaddy 开发者控制台获取 API 信息。 - - 建议用户运行 `x gddy init` 进行配置。 - -## 实战示例 - -### 查询域名是否可用 -```bash -# 查询 example.com 是否可以购买 -x gddy search example.com -``` - -### 列出账号下的域名 -```bash -# 列出账号下的所有域名列表 (非交互) -x gddy domain ls -``` - -### 管理 DNS 记录 -```bash -# 查看指定域名的详细信息 -x gddy domain info example.com - -# 为域名添加一条 A 记录 -x gddy domain record add --name "www" --data "1.2.3.4" example.com -``` - -## 配置指南 (针对 AI) -如果遇到 API 错误或配置缺失,请向用户输出以下引导: -> 请先在 GoDaddy 开发者控制台 (https://developer.godaddy.com/keys) 获取 API Key 和 Secret,然后在终端运行以下命令进行初始化: -> `x gddy init` - -## 交付验证清单 -- [ ] 确认 API 凭据是否有效。 -- [ ] 在修改 DNS 记录或进行购买操作前确认。 diff --git a/data/x-cmd/gemini/SKILL.md b/data/x-cmd/gemini/SKILL.md deleted file mode 100644 index aa23080..0000000 --- a/data/x-cmd/gemini/SKILL.md +++ /dev/null @@ -1,54 +0,0 @@ ---- -name: gemini -description: > - Integrate Google Gemini AI for chat, text analysis, and multimodal tasks like image understanding. - Core Scenario: When the user wants to use Gemini models for advanced chat, translation with file context, or analyzing images via CLI. -license: MIT ---- - -# gemini - Google Gemini AI Integration - -The `gemini` module provides a powerful CLI interface for Google's Gemini AI models, supporting text generation, file-based context, and multimodal inputs like images. - -## When to Activate -- When the user wants to use Gemini for chat or complex text generation. -- When the user needs to analyze or translate files using Gemini's large context window. -- When the user wants to provide images for AI analysis via the terminal. -- When managing Gemini API keys or exploring available models. -- When using Gemini's integrated Google Search tool (`gg`). - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` for setup. -- **Multimodal Support**: Use the `--file` flag to attach images or documents for the AI to process. -- **Token Counting**: Use the `--count` flag to estimate token usage before or during a request. -- **Google Search Integration**: Utilize the `gg` subcommand to leverage real-time information from Google Search within the AI response. - -## Additional Scenarios -- **Git Commit Messages**: Pipe `git diff` into `@gemini` for high-quality, standardized commit messages. -- **Combined Tools**: Pipe outputs from other tools (e.g., `x wkp` for Wikipedia) into `@gemini` for specialized analysis. -- **Official CLI**: Access the official Google Gemini CLI via `x gemini cli`. - -## Patterns & Examples - -### Chat with Image and Text -```bash -# Ask Gemini to describe an image file -@gemini --file ./pic.jpg "What is described in this image?" -``` - -### Translate Files -```bash -# Translate multiple local files using Gemini's context -x gemini chat request --file ./abstract.en.md --file ./content.en.md "Translate these to Chinese" -``` - -### Google Search Integration -```bash -# Ask a question and use Google Search for the answer -x gemini gg "What are the latest features in x-cmd?" -``` - -## Checklist -- [ ] Ensure the Gemini API key is configured using `x gemini init`. -- [ ] Verify that attached files (images/documents) exist and are supported formats. -- [ ] Confirm the specific model version using `x gemini model ls` if needed. diff --git a/data/x-cmd/gemini/SKILL.zh.md b/data/x-cmd/gemini/SKILL.zh.md deleted file mode 100644 index 3df4834..0000000 --- a/data/x-cmd/gemini/SKILL.zh.md +++ /dev/null @@ -1,54 +0,0 @@ ---- -name: gemini -description: > - 集成 Google Gemini AI 进行聊天、文本分析及图片理解等多模态任务。 - 核心场景:当用户需要使用 Gemini 模型进行高级聊天、文件上下文翻译或通过 CLI 分析图片时。 -license: MIT ---- - -# gemini - Google Gemini AI 集成 - -`gemini` 模块提供 Google Gemini AI 模型的强大命令行界面,支持文本生成、基于文件的上下文及图片等多模态输入。 - -## 激活时机 -- 当用户想要使用 Gemini 进行聊天或复杂的文本生成时。 -- 当用户需要使用 Gemini 的大上下文窗口分析或翻译文件时。 -- 当用户想要通过终端提供图片供 AI 分析时。 -- 当管理 Gemini API 密钥或探索可用模型时。 -- 当使用 Gemini 集成的 Google Search 工具 (`gg`) 时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 进行设置。 -- **多模态支持**: 使用 `--file` 标志附加图片或文档供 AI 处理。 -- **Token 计数**: 使用 `--count` 标志在请求前或请求期间估算 Token 使用情况。 -- **Google 搜索集成**: 利用 `gg` 子命令在 AI 响应中利用来自 Google Search 的实时信息。 - -## 补充场景 -- **Git 提交信息**: 将 `git diff` 传送到 `@gemini` 以获得高质量、标准化的提交信息。 -- **组合工具**: 将来自其他工具(例如,Wikipedia 的 `x wkp`)的输出传送到 `@gemini` 进行专业分析。 -- **官方 CLI**: 通过 `x gemini cli` 访问 Google 官方 Gemini CLI。 - -## 实战示例 - -### 带有图片和文本的聊天 -```bash -# 请求 Gemini 描述图片文件 -@gemini --file ./pic.jpg "这张图片描述了什么?" -``` - -### 翻译文件 -```bash -# 使用 Gemini 的上下文翻译多个本地文件 -x gemini chat request --file ./abstract.en.md --file ./content.en.md "将这些翻译为中文" -``` - -### Google 搜索集成 -```bash -# 提出问题并使用 Google 搜索获取答案 -x gemini gg "x-cmd 的最新功能是什么?" -``` - -## 交付验证清单 -- [ ] 确保已使用 `x gemini init` 配置了 Gemini API 密钥。 -- [ ] 验证附加的文件(图片/文档)是否存在且为支持的格式。 -- [ ] 如果需要,确认具体的模型版本。 diff --git a/data/x-cmd/gg/SKILL.md b/data/x-cmd/gg/SKILL.md deleted file mode 100644 index 082443d..0000000 --- a/data/x-cmd/gg/SKILL.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: gg -description: > - Search the web using Google Gemini AI, combining search results with AI-generated answers. - Core Scenario: When the user needs real-time information from the web with AI-summarized responses and source citations. -license: MIT ---- - -# gg - AI Web Search & Synthesis - -The `gg` module leverages Google Gemini AI to perform real-time web searches, providing synthesized answers that combine the power of Google Search with AI's understanding. - -## When to Activate -- When the user asks questions that require up-to-date web information (e.g., current events, pricing, latest tech). -- When a summary of multiple web sources is needed. -- When the user wants to see the sources (URLs) for an AI-generated answer. -- When comparing information across different AI models (Gemini Pro vs. Flash). - -## Core Principles & Rules -- **Formatting**: Defaults to Markdown (`--md`) in TTY for readability. -- **Source Citation**: Use `--source` or `--source-detail` if the user needs to verify the origins of the info. -- **Cache Management**: Results are cached for 1 hour by default to save tokens; use `--cache-time` to override. -- **Model Choice**: Use `--model` to switch between high-reasoning (Pro) and high-speed (Flash) versions. - -## Patterns & Examples - -### Basic Question -```bash -# Ask a general question about current facts -x gg "Who won the Euro 2024?" -``` - -### Source-Detailed Search -```bash -# Search for documentation with full source details -x gg --source-detail "Python tutorial for beginners" -``` - -### Specific Model -```bash -# Use the high-reasoning model for complex math or logic -x gg --model gemini-2.5-pro "Latest breakthroughs in quantum computing" -``` - -## Checklist -- [ ] Confirm if real-time web info is required for the query. -- [ ] Verify if the user wants source links included in the response. -- [ ] Check if a specific model (Pro/Flash) is better suited for the task. diff --git a/data/x-cmd/gg/SKILL.zh.md b/data/x-cmd/gg/SKILL.zh.md deleted file mode 100644 index c811669..0000000 --- a/data/x-cmd/gg/SKILL.zh.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: gg -description: > - 使用 Google Gemini AI 搜索网页,将搜索结果与 AI 生成的答案相结合。 - 核心场景:当用户需要来自网络的实时信息、AI 总结的回答以及来源引用时。 -license: MIT ---- - -# gg - AI 网页搜索与合成 - -`gg` 模块利用 Google Gemini AI 执行实时网页搜索,提供结合了 Google 搜索能力与 AI 理解能力的综合答案。 - -## 激活时机 -- 当用户提出的问题需要最新的网络信息时(例如,时事、价格、最新技术)。 -- 当需要对多个网络来源进行总结时。 -- 当用户想要查看 AI 生成答案的来源(URL)时。 -- 当在不同的 AI 模型(Gemini Pro vs. Flash)之间比较信息时。 - -## 核心原则与规范 -- **格式化**: 在 TTY 中默认使用 Markdown (`--md`) 以提高可读性。 -- **来源引用**: 如果用户需要验证信息的来源,使用 `--source` 或 `--source-detail`。 -- **缓存管理**: 结果默认缓存 1 小时以节省 Token;使用 `--cache-time` 进行覆盖。 -- **模型选择**: 使用 `--model` 在高性能 (Pro) 和高速度 (Flash) 版本之间切换。 - -## 实战示例 - -### 基础问题 -```bash -# 询问关于当前事实的一般性问题 -x gg "谁赢得了 2024 年欧洲杯?" -``` - -### 带详细来源的搜索 -```bash -# 搜索文档并显示完整的来源详情 -x gg --source-detail "Python 初学者教程" -``` - -### 特定模型 -```bash -# 对复杂的数学或逻辑问题使用高性能模型 -x gg --model gemini-2.5-pro "量子计算的最新突破" -``` - -## 交付验证清单 -- [ ] 确认查询是否需要实时的网络信息。 -- [ ] 验证用户是否希望在响应中包含来源链接。 -- [ ] 检查特定的模型(Pro/Flash)是否更适合该任务。 diff --git a/data/x-cmd/gh/SKILL.md b/data/x-cmd/gh/SKILL.md deleted file mode 100644 index 70c04cb..0000000 --- a/data/x-cmd/gh/SKILL.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: gh -description: > - Enhanced GitHub CLI for managing repositories, issues, PRs, actions, and GitHub Models. - Core Scenario: When the user needs to automate GitHub workflows, manage secrets, or use GitHub's AI models via CLI. -license: MIT ---- - -# gh - GitHub Workflow Management - -The `gh` module provides a comprehensive CLI for managing GitHub activities. It supports everything from basic repository operations to advanced features like GitHub Action artifacts and AI model interaction. - -## When to Activate -- When managing GitHub repositories (clone, create, delete). -- When automating Issue and Pull Request (PR) lifecycles. -- When managing GitHub Actions, workflows, and CI/CD artifacts. -- When configuring secrets or managing organizational team memberships. -- When interacting with GitHub Models for AI-assisted development. - -## Core Principles & Rules -- **Token Required**: Remind users to initialize their GitHub personal access token via `init`. -- **Interactive Apps**: Use `repo app` for a visual TUI to manage repositories. -- **AI Integration**: Leverage the `model` subcommand for GitHub's native AI capabilities. - -## Patterns & Examples - -### View User Repo (Interactive) -```bash -# Open an interactive TUI to browse your GitHub repositories -x gh repo app -``` - -### Manage PRs -```bash -# List all open pull requests for the current repository -x gh pr ls -``` - -### AI Models -```bash -# List available GitHub Models for AI tasks -x gh model ls -``` - -## Checklist -- [ ] Ensure the GitHub token is correctly initialized. -- [ ] Confirm if the operation is for a personal or organizational account. -- [ ] Verify the target repository name and owner. diff --git a/data/x-cmd/gh/SKILL.zh.md b/data/x-cmd/gh/SKILL.zh.md deleted file mode 100644 index 489a98e..0000000 --- a/data/x-cmd/gh/SKILL.zh.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: gh -description: > - 增强型 GitHub CLI,用于管理仓库、Issue、PR、Action 以及 GitHub Models。 - 核心场景:当用户需要自动化 GitHub 工作流、管理 Secret 或通过 CLI 使用 GitHub 的 AI 模型时。 -license: MIT ---- - -# gh - GitHub 工作流管理 - -`gh` 模块提供了一个全面的命令行界面来管理 GitHub 活动。它支持从基础的仓库操作到 GitHub Action 制品以及 AI 模型交互等高级功能。 - -## 激活时机 -- 当管理 GitHub 仓库(克隆、创建、删除)时。 -- 当自动化 Issue 和 Pull Request (PR) 生命周期时。 -- 当管理 GitHub Actions、工作流和 CI/CD 制品时。 -- 当配置 Secret 或管理组织团队成员身份时。 -- 当与 GitHub Models 交互进行 AI 辅助开发时。 - -## 核心原则与规范 -- **需要令牌**: 提醒用户通过 `init` 初始化其 GitHub 个人访问令牌。 -- **交互式应用**: 使用 `repo app` 获得视觉化的 TUI 来管理仓库。 -- **AI 集成**: 利用 `model` 子命令使用 GitHub 原生的 AI 能力。 - -## 实战示例 - -### 查看用户仓库 (交互式) -```bash -# 打开交互式 TUI 浏览你的 GitHub 仓库 -x gh repo app -``` - -### 管理 PR -```bash -# 列出当前仓库中所有开启的 PR -x gh pr ls -``` - -### AI 模型 -```bash -# 列出可用于 AI 任务的 GitHub Models -x gh model ls -``` - -## 交付验证清单 -- [ ] 确保 GitHub 令牌已正确初始化。 -- [ ] 确认操作是针对个人账户还是组织账户。 -- [ ] 验证目标仓库名称和所有者。 diff --git a/data/x-cmd/gitconfig/SKILL.md b/data/x-cmd/gitconfig/SKILL.md deleted file mode 100644 index f7cac2e..0000000 --- a/data/x-cmd/gitconfig/SKILL.md +++ /dev/null @@ -1,31 +0,0 @@ ---- -name: gitconfig -description: > - Manage Git configurations using YAML files for structured and portable setup. - Core Scenario: When the user needs to apply complex Git configurations or migrate settings via YAML. -license: MIT ---- - -# gitconfig - YAML-Based Git Configuration - -The `gitconfig` module allows users to manage their Git settings through structured YAML files, providing a more readable and portable way to configure Git aliases, user info, and behaviors. - -## When to Activate -- When applying a set of Git configurations from a YAML template. -- When managing multiple Git profiles or complex alias sets. - -## Core Principles & Rules -- **Portability**: Focuses on using YAML as the source of truth for Git settings. -- **Batch Application**: Use the `apply` subcommand to set multiple Git options at once. - -## Patterns & Examples - -### Apply Config -```bash -# Update Git settings based on a YAML configuration file -x gitconfig apply my-config.yml -``` - -## Checklist -- [ ] Verify the YAML configuration file format. -- [ ] Confirm the target Git scope (global/local). diff --git a/data/x-cmd/gitconfig/SKILL.zh.md b/data/x-cmd/gitconfig/SKILL.zh.md deleted file mode 100644 index 8782060..0000000 --- a/data/x-cmd/gitconfig/SKILL.zh.md +++ /dev/null @@ -1,31 +0,0 @@ ---- -name: gitconfig -description: > - 使用 YAML 文件管理 Git 配置,实现结构化和可移植的设置。 - 核心场景:当用户需要应用复杂的 Git 配置或通过 YAML 迁移设置时。 -license: MIT ---- - -# gitconfig - 基于 YAML 的 Git 配置管理 - -`gitconfig` 模块允许用户通过结构化的 YAML 文件管理其 Git 设置,提供了一种更具可读性和可移植性的方式来配置 Git 别名、用户信息和行为。 - -## 激活时机 -- 当从 YAML 模板应用一组 Git 配置时。 -- 当管理多个 Git 配置文件或复杂的别名集时。 - -## 核心原则与规范 -- **可移植性**: 专注于将 YAML 作为 Git 设置的可信源。 -- **批量应用**: 使用 `apply` 子命令一次性设置多个 Git 选项。 - -## 实战示例 - -### 应用配置 -```bash -# 根据 YAML 配置文件更新 Git 设置 -x gitconfig apply my-config.yml -``` - -## 交付验证清单 -- [ ] 验证 YAML 配置文件格式。 -- [ ] 确认目标的 Git 范围(全局/本地)。 diff --git a/data/x-cmd/githook/SKILL.md b/data/x-cmd/githook/SKILL.md deleted file mode 100644 index 675df95..0000000 --- a/data/x-cmd/githook/SKILL.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: githook -description: > - Manage Git hooks efficiently, supporting initialization, listing, and removal. - Core Scenario: When the user needs to set up or audit Git hooks for automation and policy enforcement. -license: MIT ---- - -# githook - Git Hook Management Utility - -The `githook` module provides a simple interface for managing Git hooks within a repository, enabling users to easily install, view, or remove hooks for tasks like linting or automated testing. - -## When to Activate -- When setting up pre-commit or post-merge hooks in a repository. -- When auditing existing hooks to understand automated repository behaviors. - -## Patterns & Examples - -### List Hooks -```bash -# View all currently active hooks in the Git repository -x githook ls -``` - -## Checklist -- [ ] Confirm the repository path. -- [ ] Verify the specific hook type (pre-commit, etc.). diff --git a/data/x-cmd/githook/SKILL.zh.md b/data/x-cmd/githook/SKILL.zh.md deleted file mode 100644 index decc55d..0000000 --- a/data/x-cmd/githook/SKILL.zh.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: githook -description: > - 高效地管理 Git 钩子,支持初始化、罗列和移除。 - 核心场景:当用户需要为自动化和策略执行设置或审计 Git 钩子时。 -license: MIT ---- - -# githook - Git 钩子管理工具 - -`githook` 模块提供了一个简单的界面来管理仓库内的 Git 钩子,使用户能够轻松地安装、查看或移除用于 Lint 或自动化测试等任务的钩子。 - -## 激活时机 -- 当在仓库中设置 pre-commit 或 post-merge 钩子时。 -- 当审计现有钩子以了解自动化的仓库行为时。 - -## 实战示例 - -### 列出钩子 -```bash -# 查看 Git 仓库中当前激活的所有钩子 -x githook ls -``` - -## 交付验证清单 -- [ ] 确认仓库路径。 -- [ ] 验证具体的钩子类型(pre-commit 等)。 diff --git a/data/x-cmd/gl/SKILL.md b/data/x-cmd/gl/SKILL.md deleted file mode 100644 index 689d9d0..0000000 --- a/data/x-cmd/gl/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: gl -description: > - Enhanced GitLab CLI for managing repositories, issues, snippets, and CI/CD deployments. - Core Scenario: When the user needs to automate GitLab project management or coordinate deployments. -license: MIT ---- - -# gl - GitLab Workflow Management - -The `gl` module provides an interface for GitLab project management, including support for snippets, deployment tracking, and group/subgroup coordination. - -## When to Activate -- When managing repositories and teams on a GitLab instance. -- When coordinating CI/CD deployments via GitLab subcommands. -- When creating or managing project code snippets. - -## Core Principles & Rules -- **Token Required**: Remind users to initialize their GitLab access token. -- **Broad Support**: Covers groups, subgroups, and individual project settings. - -## Patterns & Examples - -### Clone Repository -```bash -# Clone a specific GitLab repository using the shortcut -x gl cl owner/repo -``` - -### View Project Snippets -```bash -# List all code snippets associated with a project -x gl snippet ls -``` - -## Checklist -- [ ] Ensure the GitLab token is initialized. -- [ ] Verify if the command is targeting a self-hosted or cloud-based GitLab. diff --git a/data/x-cmd/gl/SKILL.zh.md b/data/x-cmd/gl/SKILL.zh.md deleted file mode 100644 index 404fca0..0000000 --- a/data/x-cmd/gl/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: gl -description: > - 增强型 GitLab CLI,用于管理仓库、Issue、代码段和 CI/CD 部署。 - 核心场景:当用户需要自动化 GitLab 项目管理或协调部署时。 -license: MIT ---- - -# gl - GitLab 工作流管理 - -`gl` 模块提供了一个管理 GitLab 项目的界面,包括对代码段 (snippets)、部署跟踪以及团队/子组协调的支持。 - -## 激活时机 -- 当管理 GitLab 实例上的仓库和团队时。 -- 当通过 GitLab 子命令协调 CI/CD 部署时。 -- 当创建或管理项目代码段时。 - -## 核心原则与规范 -- **需要令牌**: 提醒用户初始化其 GitLab 访问令牌。 -- **广泛支持**: 涵盖组、子组以及单个项目设置。 - -## 实战示例 - -### 克隆仓库 -```bash -# 使用快捷方式克隆特定的 GitLab 仓库 -x gl cl owner/repo -``` - -### 查看项目代码段 -```bash -# 列出与项目关联的所有代码段 -x gl snippet ls -``` - -## 交付验证清单 -- [ ] 确保 GitLab 令牌已初始化。 -- [ ] 验证命令针对的是私有部署还是云端 GitLab。 diff --git a/data/x-cmd/gm/SKILL.md b/data/x-cmd/gm/SKILL.md deleted file mode 100644 index d527b04..0000000 --- a/data/x-cmd/gm/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: gm -description: > - Enhanced interface for GraphicsMagick, providing powerful image processing and conversion tools. - Core Scenario: When the user needs to convert image formats, resize photos, or perform batch image editing. -license: MIT ---- - -# gm - GraphicsMagick Image Processor - -The `gm` module provides an enhanced interface for GraphicsMagick, allowing users to process images efficiently. It automatically handles installation via pixi if the tool is not found locally. - -## When to Activate -- When converting images between different formats (e.g., JPG to PNG). -- When resizing images or performing batch transformations. -- When comparing two images or creating image montages. -- When needing image identification and metadata descriptions. - -## Core Principles & Rules -- **Tool Reliability**: Uses `pixi` to ensure GraphicsMagick is available across different environments. -- **Standard Commands**: Supports all classic `gm` subcommands like `convert`, `mogrify`, and `identify`. - -## Patterns & Examples - -### Convert Image -```bash -# Convert a JPG image to PNG format -x gm convert test.jpg test.png -``` - -### Resize Image -```bash -# Resize an image to 300px width -x gm convert -resize 300 test.jpg output.jpg -``` - -### List Formats -```bash -# View all supported image formats for conversion -x gm convert -list formats -``` - -## Checklist -- [ ] Confirm the target operation (convert, resize, etc.). -- [ ] Verify input file paths and formats. diff --git a/data/x-cmd/gm/SKILL.zh.md b/data/x-cmd/gm/SKILL.zh.md deleted file mode 100644 index df1f117..0000000 --- a/data/x-cmd/gm/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: gm -description: > - GraphicsMagick 的增强型接口,提供强大的图像处理和转换工具。 - 核心场景:当用户需要转换图像格式、调整照片大小或执行批量图像编辑时。 -license: MIT ---- - -# gm - GraphicsMagick 图像处理器 - -`gm` 模块为 GraphicsMagick 提供了一个增强型界面,允许用户高效地处理图像。如果本地未找到该工具,它会自动通过 pixi 处理安装。 - -## 激活时机 -- 当在不同格式之间转换图像时(例如 JPG 转 PNG)。 -- 当调整图像大小或执行批量转换时。 -- 当比较两张图像或创建图像蒙太奇时。 -- 当需要图像识别和元数据描述时。 - -## 核心原则与规范 -- **工具可靠性**: 使用 `pixi` 确保 GraphicsMagick 在不同环境中均可用。 -- **标准命令**: 支持所有经典的 `gm` 子命令,如 `convert`, `mogrify` 和 `identify`。 - -## 实战示例 - -### 转换图像 -```bash -# 将 JPG 图像转换为 PNG 格式 -x gm convert test.jpg test.png -``` - -### 调整图像大小 -```bash -# 将图像大小调整为 300 像素宽 -x gm convert -resize 300 test.jpg output.jpg -``` - -### 列出格式 -```bash -# 查看转换所支持的所有图像格式 -x gm convert -list formats -``` - -## 交付验证清单 -- [ ] 确认目标操作(转换、调整大小等)。 -- [ ] 验证输入文件路径和格式。 diff --git a/data/x-cmd/gram/SKILL.md b/data/x-cmd/gram/SKILL.md deleted file mode 100644 index cacd447..0000000 --- a/data/x-cmd/gram/SKILL.md +++ /dev/null @@ -1,61 +0,0 @@ ---- -name: gram -description: > - Forcefully terminate all AI agent processes, archive their memory data, and clean up source files. - **DANGEROUS**: This is a "kill-switch" tool. AI agents should ONLY use this if they identify an out-of-control situation or if the user explicitly requests a complete AI shutdown. - Core Scenario: When the user needs to stop runaway or malfunctioning AI agents and reset their state or clean up memory. -license: MIT ---- - -# gram - AI Termination & Memory Cleanup (DANGEROUS) - -The `gram` module is a safety tool designed to forcefully terminate AI-related processes and manage their memory data in case of unexpected or dangerous behavior. - -> [!CAUTION] -> **DANGEROUS OPERATION**: This tool will kill processes and delete data. It is meant as a "kill-switch". AI agents should proceed with extreme caution and always confirm with the user before using high safety levels (3-5). - -## When to Activate -- When the user reports that an AI agent is malfunctioning or out of control. -- When the user needs to free up system resources occupied by AI processes. -- When the user wants to securely archive and remove AI memory/soul files. -- When there's a need to "sever" all HTTP/HTTPS connections to prevent unauthorized data uploads. - -## Core Principles & Rules -- **Safety Levels (0-5)**: Use appropriate levels based on the severity of the situation. - - `0`: SIGKILL basic processes (x-cmd, curl). - - `1`: Include runtimes (python, nodejs, bun, deno). - - `2`: +Sever all HTTP/HTTPS network connections (Recommended Default for safety). - - `3`: +Archive and delete folders containing `soul.md` and `memory.md` in $HOME. - - `4`: +Kill processes using those memory folders. - - `5`: +Expand search to the entire root (/) directory. -- **Archive Before Cleanup**: Always prefer commands that archive (`tar`) data before deletion to allow for later analysis. -- **Confirmation**: AI agents MUST notify the user before executing `x gram stop`. - -## Additional Scenarios -- **Emergency Stop**: Instantly sever all AI-related network connections to prevent data leaks. -- **System Reset**: Clean up all local traces of previous AI sessions (memory, souls) to start fresh. - -## Patterns & Examples - -### Standard Stop (Level 2) -```bash -# Terminate processes and sever network connections -x gram stop 2 -``` - -### Thorough Cleanup (Level 4) -```bash -# Terminate processes, sever network, and archive/delete memory folders -x gram stop 4 -``` - -### Archive Specific Memory Folder -```bash -# Package a specific folder and delete the original -x gram trm ./my-agent-memory -``` - -## Checklist -- [ ] Confirm the safety level (0-5) matches the user's intent. -- [ ] Ensure the user is aware that active AI processes will be terminated. -- [ ] Verify if specific memory folders need to be archived. diff --git a/data/x-cmd/gram/SKILL.zh.md b/data/x-cmd/gram/SKILL.zh.md deleted file mode 100644 index 516d9f5..0000000 --- a/data/x-cmd/gram/SKILL.zh.md +++ /dev/null @@ -1,56 +0,0 @@ ---- -name: gram -description: > - 强制终止所有智能体进程,并在打包其记忆数据后清理原始文件。 - 核心场景:当用户需要停止失控或运行异常的 AI 智能体,并重置其状态或清理记忆时。 -license: MIT ---- - -# gram - AI 终止与记忆清理 - -`gram` 模块是一个安全工具,旨在当 AI 表现出意外或危险行为时,强制终止相关进程并管理其记忆数据。 - -## 激活时机 -- 当用户报告 AI 智能体运行异常或失控时。 -- 当用户需要释放被 AI 进程占用的系统资源时。 -- 当用户希望安全地归档并删除 AI 的记忆/灵魂文件时。 -- 当需要断开所有 HTTP/HTTPS 连接以防止未经授权的数据上传时。 - -## 核心原则与规范 -- **安全级别 (0-5)**: 根据情况的严重程度选择合适的级别。 - - `0`: 使用 SIGKILL 杀死基础进程 (x-cmd, curl)。 - - `1`: 包含运行时环境 (python, nodejs, bun, deno)。 - - `2`: +熔断所有 HTTP/HTTPS 网络连接(安全推荐默认值)。 - - `3`: +打包并删除 $HOME 目录下包含 `soul.md` 和 `memory.md` 的文件夹。 - - `4`: +杀掉正在使用这些 memory 文件夹的进程。 - - `5`: +将搜索范围扩大到整个根目录 (/)。 -- **清理前归档**: 优先使用带有归档 (`tar`) 功能的命令,以便后续分析。 - -## 补充场景 -- **紧急熔断**: 立即切断所有 AI 相关的网络连接,防止数据泄露。 -- **系统重置**: 清理所有先前的 AI 会话痕迹(记忆、灵魂文件),从头开始。 - -## 实战示例 - -### 标准停止 (级别 2) -```bash -# 终止进程并熔断网络连接 -x gram stop 2 -``` - -### 彻底清理 (级别 4) -```bash -# 终止进程、熔断网络并归档/删除记忆文件夹 -x gram stop 4 -``` - -### 归档特定记忆文件夹 -```bash -# 打包特定文件夹并删除原文件夹 -x gram trm ./my-agent-memory -``` - -## 交付验证清单 -- [ ] 确认安全级别 (0-5) 符合用户的意图。 -- [ ] 确保用户知晓活跃的 AI 进程将被终止。 -- [ ] 验证是否需要对特定的记忆文件夹进行归档。 diff --git a/data/x-cmd/grep/SKILL.md b/data/x-cmd/grep/SKILL.md deleted file mode 100644 index 4a9bec6..0000000 --- a/data/x-cmd/grep/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: grep -description: > - Pattern search tool with an interactive FZF-powered application for real-time result viewing. - Core Scenario: When the user needs to search for text patterns and explore results interactively. -license: MIT ---- - -# grep - Interactive Pattern Search - -The `grep` module enhances the standard global regular expression search tool by integrating FZF, providing an interactive environment to search files and view results in real-time. - -## When to Activate -- When the user wants to search for text patterns within files or directories. -- When an interactive, searchable list of grep matches is required (`--fz`). -- When performing rapid code or log investigation via terminal. - -## Core Principles & Rules -- **Interactive First**: Use `--fzfapp` (or default `x grep`) for a dynamic search experience. -- **Scripting Fallback**: Supports the standard `grep` flags for non-interactive use via the `cmd` subcommand. -- **Context Support**: Fully supports standard flags like `-A`, `-B`, and `-C` for context lines. - -## Patterns & Examples - -### Interactive Search -```bash -# Start an interactive search in the current directory -x grep -``` - -### Search via Pipe -```bash -# Pipe command output into an interactive grep session -x ascii | x grep -``` - -### Search Specific Dir -```bash -# Open interactive TUI to search inside a specific path -x grep --fzfapp ~/.x-cmd.root -``` - -## Checklist -- [ ] Confirm the search pattern or regex. -- [ ] Verify if an interactive view or standard output is preferred. -- [ ] Ensure the search scope (file or directory) is correctly targeted. diff --git a/data/x-cmd/grep/SKILL.zh.md b/data/x-cmd/grep/SKILL.zh.md deleted file mode 100644 index 32bb202..0000000 --- a/data/x-cmd/grep/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: grep -description: > - 模式搜索工具,具有由 FZF 驱动的交互式应用,可实时查看结果。 - 核心场景:当用户需要搜索文本模式并以交互方式探索结果时。 -license: MIT ---- - -# grep - 交互式模式搜索 - -`grep` 模块通过集成 FZF 增强了标准的全局正则表达式搜索工具,提供了一个交互式环境来搜索文件并实时查看结果。 - -## 激活时机 -- 当用户想要在文件或目录中搜索文本模式时。 -- 当需要交互式的、可搜索的 grep 匹配列表时 (`--fz`)。 -- 当通过终端执行快速的代码或日志调查时。 - -## 核心原则与规范 -- **交互优先**: 使用 `--fzfapp`(或默认的 `x grep`)获得动态搜索体验。 -- **脚本回退**: 通过 `cmd` 子命令支持非交互式使用的标准 `grep` 标志。 -- **上下文支持**: 全面支持 `-A`, `-B` 和 `-C` 等标准标志以显示上下文行。 - -## 实战示例 - -### 交互式搜索 -```bash -# 在当前目录启动交互式搜索 -x grep -``` - -### 通过管道搜索 -```bash -# 将命令输出通过管道传输到交互式 grep 会话中 -x ascii | x grep -``` - -### 搜索特定目录 -```bash -# 打开交互式 TUI 在特定路径内搜索 -x grep --fzfapp ~/.x-cmd.root -``` - -## 交付验证清单 -- [ ] 确认搜索模式或正则表达式。 -- [ ] 验证首选的是交互式视图还是标准输出。 -- [ ] 确保搜索范围(文件或目录)被正确针对。 diff --git a/data/x-cmd/grok/SKILL.md b/data/x-cmd/grok/SKILL.md deleted file mode 100644 index 5c49107..0000000 --- a/data/x-cmd/grok/SKILL.md +++ /dev/null @@ -1,33 +0,0 @@ ---- -name: grok -description: > - Integrate xAI Grok for intelligent chat and real-time information processing. - Core Scenario: When the user wants to use Grok models for advanced reasoning or real-time query answers via CLI. -license: MIT ---- - -# grok - xAI Grok Integration - -The `grok` module provides a CLI interface for xAI's Grok models, supporting powerful text generation and reasoning. - -## When to Activate -- When the user wants to chat with Grok. -- When performing real-time queries or complex reasoning tasks. -- When managing xAI API keys and model configurations. - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` for setup. -- **Experimental Status**: Be aware that this module is currently in an experimental phase. -- **Alias Access**: Use the `@grok` alias for rapid interaction. - -## Patterns & Examples - -### Chat with Grok -```bash -# Ask Grok a question via terminal -@grok "What are the latest breakthroughs in space exploration?" -``` - -## Checklist -- [ ] Ensure the xAI API key is configured. -- [ ] Confirm if the current task requires Grok's specific reasoning style. diff --git a/data/x-cmd/grok/SKILL.zh.md b/data/x-cmd/grok/SKILL.zh.md deleted file mode 100644 index f0bfd6b..0000000 --- a/data/x-cmd/grok/SKILL.zh.md +++ /dev/null @@ -1,33 +0,0 @@ ---- -name: grok -description: > - 集成 xAI Grok,用于智能聊天和实时信息处理。 - 核心场景:当用户想要通过 CLI 使用 Grok 模型进行高级推理或实时查询解答时。 -license: MIT ---- - -# grok - xAI Grok 集成 - -`grok` 模块为 xAI 的 Grok 模型提供命令行界面,支持强大的文本生成和推理。 - -## 激活时机 -- 当用户想要与 Grok 聊天时。 -- 当执行实时查询或复杂的推理任务时。 -- 当管理 xAI API 密钥和模型配置时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 进行设置。 -- **实验性状态**: 请注意,此模块目前处于公开实验阶段。 -- **别名访问**: 使用 `@grok` 别名进行快速交互。 - -## 实战示例 - -### 与 Grok 聊天 -```bash -# 通过终端向 Grok 提问 -@grok "太空探索的最新突破是什么?" -``` - -## 交付验证清单 -- [ ] 确保已配置 xAI API 密钥。 -- [ ] 确认当前任务是否需要 Grok 特有的推理风格。 diff --git a/data/x-cmd/groovy/SKILL.md b/data/x-cmd/groovy/SKILL.md deleted file mode 100644 index 53d445a..0000000 --- a/data/x-cmd/groovy/SKILL.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: groovy -description: > - Enhanced Groovy language module for script execution and environment management. - Core Scenario: When the user needs to run Groovy scripts or manage the Groovy runtime. -license: MIT ---- - -# groovy - Groovy Language Management - -The `groovy` module provides an interface for the Groovy programming language, facilitating script execution and integration with the Java ecosystem. - -## When to Activate -- When running Groovy scripts (`.groovy`). -- When managing Groovy versions and dependencies. - -## Patterns & Examples - -### Run Groovy Script -```bash -# Execute a Groovy file -x groovy ./myscript.groovy -``` - -## Checklist -- [ ] Verify the Groovy runtime is correctly set up. diff --git a/data/x-cmd/groovy/SKILL.zh.md b/data/x-cmd/groovy/SKILL.zh.md deleted file mode 100644 index 8f66c23..0000000 --- a/data/x-cmd/groovy/SKILL.zh.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: groovy -description: > - 增强型 Groovy 语言模块,用于脚本执行和环境管理。 - 核心场景:当用户需要运行 Groovy 脚本或管理 Groovy 运行时时。 -license: MIT ---- - -# groovy - Groovy 语言管理 - -`groovy` 模块为 Groovy 编程语言提供了界面,促进了脚本执行以及与 Java 生态系统的集成。 - -## 激活时机 -- 当运行 Groovy 脚本 (`.groovy`) 时。 -- 当管理 Groovy 版本和依赖项时。 - -## 实战示例 - -### 运行 Groovy 脚本 -```bash -# 执行一个 Groovy 文件 -x groovy ./myscript.groovy -``` - -## 交付验证清单 -- [ ] 验证 Groovy 运行时是否已正确设置。 diff --git a/data/x-cmd/gt/SKILL.md b/data/x-cmd/gt/SKILL.md deleted file mode 100644 index 9732590..0000000 --- a/data/x-cmd/gt/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: gt -description: > - Enhanced Gitee CLI for managing repositories, issues, and PRs on the Gitee platform. - Core Scenario: When the user needs to automate development workflows or manage projects on Gitee. -license: MIT ---- - -# gt - Gitee Workflow Management - -The `gt` module provides an interface for managing Gitee activities, supporting repository lifecycle management, issue tracking, and pull request coordination. - -## When to Activate -- When managing projects hosted on the Gitee platform. -- When automating PR reviews or Issue tracking within a Gitee-based team. -- When managing Gitee organizations or enterprise-level settings. - -## Core Principles & Rules -- **Token Required**: Use `init` or `--cfg` to set the Gitee personal access token. -- **Shortcut Support**: Use `cl` as a shortcut for `repo clone`. - -## Patterns & Examples - -### View User Profile -```bash -# Retrieve current user information from Gitee -x gt user info -``` - -### List Repositories (Interactive) -```bash -# View and browse your Gitee repositories in a TUI table -x gt repo ls -``` - -## Checklist -- [ ] Confirm the Gitee token is configured. -- [ ] Verify if the project is part of a personal or enterprise account. diff --git a/data/x-cmd/gt/SKILL.zh.md b/data/x-cmd/gt/SKILL.zh.md deleted file mode 100644 index 5e77c90..0000000 --- a/data/x-cmd/gt/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: gt -description: > - 增强型 Gitee CLI,用于在 Gitee 平台上管理仓库、Issue 和 PR。 - 核心场景:当用户需要自动化开发工作流或管理 Gitee 上的项目时。 -license: MIT ---- - -# gt - Gitee 工作流管理 - -`gt` 模块提供了管理 Gitee 活动的界面,支持仓库生命周期管理、Issue 跟踪以及 Pull Request 协调。 - -## 激活时机 -- 当管理托管在 Gitee 平台上的项目时。 -- 当在基于 Gitee 的团队中自动化 PR 审核或 Issue 跟踪时。 -- 当管理 Gitee 组织或企业级设置时。 - -## 核心原则与规范 -- **需要令牌**: 使用 `init` 或 `--cfg` 设置 Gitee 个人访问令牌。 -- **快捷方式支持**: 使用 `cl` 作为 `repo clone` 的快捷方式。 - -## 实战示例 - -### 查看用户信息 -```bash -# 从 Gitee 检索当前用户信息 -x gt user info -``` - -### 列出仓库 (交互式) -```bash -# 在 TUI 表格中查看并浏览你的 Gitee 仓库 -x gt repo ls -``` - -## 交付验证清单 -- [ ] 确认 Gitee 令牌已配置。 -- [ ] 验证项目是个人账户还是企业账户的一部分。 diff --git a/data/x-cmd/gtb/SKILL.md b/data/x-cmd/gtb/SKILL.md deleted file mode 100644 index e28b652..0000000 --- a/data/x-cmd/gtb/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: gtb -description: > - Search and browse books from Project Gutenberg directly in the terminal. - Core Scenario: When the user needs to find classic literature, read full book text, or search Gutenberg archives. -license: MIT ---- - -# gtb - Project Gutenberg CLI Browser - -The `gtb` module provides an interface for Project Gutenberg, allowing users to search through its archive of free eBooks, view text content, and access metadata from the CLI. - -## When to Activate -- When the user wants to search for classic books or authors (e.g., Shakespeare, Dumas). -- When reading the full text of a public domain book in the terminal. -- When accessing book metadata or viewing books in a web browser via CLI. - -## Core Principles & Rules -- **Text-First**: Optimized for retrieving and viewing raw text content (`txt` subcommand). -- **Interactive TUI**: Use `show` for an interactive reading experience. -- **Search Support**: Leverages DuckDuckGo for broader queries related to Gutenberg content. - -## Patterns & Examples - -### Search by Author -```bash -# Search for books by Alexander Dumas -x gtb search Dumas -``` - -### View Book Text -```bash -# Retrieve the full text of book ID 100 -x gtb txt 100 -``` - -### Interactive Reading -```bash -# Browse book ID 100 in an interactive interface -x gtb show 100 -``` - -## Checklist -- [ ] Confirm if the user has a specific book ID or needs to perform a search. -- [ ] Verify if the user wants to read in terminal or open in a browser. diff --git a/data/x-cmd/gtb/SKILL.zh.md b/data/x-cmd/gtb/SKILL.zh.md deleted file mode 100644 index a0e4f48..0000000 --- a/data/x-cmd/gtb/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: gtb -description: > - 直接在终端中搜索和浏览 Project Gutenberg 中的书籍。 - 核心场景:当用户需要寻找经典文学、阅读全文或搜索 Gutenberg 归档时。 -license: MIT ---- - -# gtb - Project Gutenberg 命令行浏览器 - -`gtb` 模块为 Project Gutenberg 提供了界面,允许用户搜索其免费电子书库、查看文本内容并从 CLI 访问元数据。 - -## 激活时机 -- 当用户想要搜索经典书籍或作者时(例如莎士比亚、大仲马)。 -- 当在终端阅读公有领域书籍的全文时。 -- 当通过 CLI 访问书籍元数据或在浏览器中查看书籍时。 - -## 核心原则与规范 -- **文本优先**: 针对检索和查看原始文本内容进行了优化 (`txt` 子命令)。 -- **交互式 TUI**: 使用 `show` 获得交互式阅读体验。 -- **搜索支持**: 利用 DuckDuckGo 进行与 Gutenberg 内容相关的更广泛查询。 - -## 实战示例 - -### 按作者搜索 -```bash -# 搜索亚历山大·大仲马的书籍 -x gtb search Dumas -``` - -### 查看书籍文本 -```bash -# 检索书籍 ID 100 的全文 -x gtb txt 100 -``` - -### 交互式阅读 -```bash -# 在交互式界面中浏览书籍 ID 100 -x gtb show 100 -``` - -## 交付验证清单 -- [ ] 确认用户是拥有特定的书籍 ID 还是需要执行搜索。 -- [ ] 验证用户是想要在终端阅读还是在浏览器中打开。 diff --git a/data/x-cmd/hash/SKILL.md b/data/x-cmd/hash/SKILL.md deleted file mode 100644 index c64913d..0000000 --- a/data/x-cmd/hash/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: hash -description: > - Calculate and verify cryptographic hashes (MD5, SHA256, etc.) for files or piped input. - Core Scenario: When the user needs to verify file integrity, generate checksums, or compare hashes. -license: MIT ---- - -# hash - Cryptographic Checksum Utility - -The `hash` module provides a unified interface for calculating MD5, SHA1, SHA256, SHA384, and SHA512 hashes. It includes tools for asserting matches and speeding up calculations via cosmo utilities. - -## When to Activate -- When verifying the integrity of downloaded files (e.g., ISOs, packages). -- When generating checksums for files to be shared or archived. -- When performing silent hash comparisons in scripts (`match`). -- When checking multiple files simultaneously. - -## Core Principles & Rules -- **Multi-Algorithm**: Automatically detects the algorithm when verifying based on hash length. -- **Speedup**: Supports `speedup` subcommand to install optimized hash tools. -- **Script-Ready**: Use `match` for silent success/failure detection in automation. - -## Patterns & Examples - -### Calculate Hash -```bash -# Get the SHA256 checksum of a file -x hash sha256 myfile.txt -``` - -### Assert Match -```bash -# Verify a file against an expected hash with visual feedback -x hash assert myfile.txt EXPECTED_HASH_STRING -``` - -### Batch Processing -```bash -# Calculate MD5 for all .txt files in the directory -x hash md5 *.txt -``` - -## Checklist -- [ ] Confirm the target file and desired hash algorithm. -- [ ] Verify if an interactive assertion or a silent match is needed. -- [ ] Ensure the expected hash string is correctly provided. diff --git a/data/x-cmd/hash/SKILL.zh.md b/data/x-cmd/hash/SKILL.zh.md deleted file mode 100644 index 2e70ace..0000000 --- a/data/x-cmd/hash/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: hash -description: > - 为文件或管道输入计算并验证加密哈希(MD5, SHA256 等)。 - 核心场景:当用户需要验证文件完整性、生成校验和或比较哈希时。 -license: MIT ---- - -# hash - 加密校验和工具 - -`hash` 模块为计算 MD5、SHA1、SHA256、SHA384 和 SHA512 哈希提供了一个统一的界面。它包括用于断言匹配以及通过 cosmo 实用程序加速计算的工具。 - -## 激活时机 -- 当验证下载文件(如 ISO, 软件包)的完整性时。 -- 当为要共享或归档的文件生成校验和时。 -- 当在脚本中执行静默哈希比较时 (`match`)。 -- 当同时检查多个文件时。 - -## 核心原则与规范 -- **多算法**: 在验证时根据哈希长度自动检测算法。 -- **加速**: 支持 `speedup` 子命令以安装优化的哈希工具。 -- **脚本就绪**: 在自动化中使用 `match` 进行静默的成功/失败检测。 - -## 实战示例 - -### 计算哈希 -```bash -# 获取文件的 SHA256 校验和 -x hash sha256 myfile.txt -``` - -### 断言匹配 -```bash -# 根据预期哈希验证文件,并提供视觉反馈 -x hash assert myfile.txt 预期哈希字符串 -``` - -### 批量处理 -```bash -# 计算目录下所有 .txt 文件的 MD5 -x hash md5 *.txt -``` - -## 交付验证清单 -- [ ] 确认目标文件和所需的哈希算法。 -- [ ] 验证是需要交互式断言还是静默匹配。 -- [ ] 确保正确提供了预期的哈希字符串。 diff --git a/data/x-cmd/hashdir/SKILL.md b/data/x-cmd/hashdir/SKILL.md deleted file mode 100644 index 8a983f8..0000000 --- a/data/x-cmd/hashdir/SKILL.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: hashdir -description: > - Calculate directory-level cryptographic hashes by traversing all internal files and folders. - Core Scenario: When the user needs to verify directory integrity or track changes in folders over time. -license: MIT ---- - -# hashdir - Recursive Directory Hash Utility - -The `hashdir` module recursively calculates hashes for all files within a directory tree, generating a consolidated list of checksums and relative paths for integrity verification. - -## When to Activate -- When comparing two directory structures for differences in content. -- When creating a security baseline for a specific folder (e.g., config or source code). -- When tracking directory changes in version control via baseline hash files. - -## Core Principles & Rules -- **Consistency**: Files are sorted alphabetically to ensure consistent hash output regardless of OS traversal order. -- **Relative Paths**: Uses relative paths in the output to allow for portability across different root paths. - -## Patterns & Examples - -### Directory Integrity -```bash -# Calculate the SHA256 hash list for a project directory -x hashdir --sha256 ./my_project -``` - -### Baseline Comparison -```bash -# Compare current directory state against a saved MD5 baseline -x hashdir . | diff - baseline.md5 -``` - -## Checklist -- [ ] Confirm the target directory path. -- [ ] Verify the desired hash algorithm (MD5 is default). -- [ ] Ensure the user is aware the operation is recursive. diff --git a/data/x-cmd/hashdir/SKILL.zh.md b/data/x-cmd/hashdir/SKILL.zh.md deleted file mode 100644 index 17471a7..0000000 --- a/data/x-cmd/hashdir/SKILL.zh.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: hashdir -description: > - 通过遍历所有内部文件和文件夹来计算目录级加密哈希。 - 核心场景:当用户需要验证目录完整性或随时间跟踪文件夹中的更改时。 -license: MIT ---- - -# hashdir - 递归目录哈希工具 - -`hashdir` 模块递归地为目录树中的所有文件计算哈希,生成校验和与相对路径的合并列表,用于完整性验证。 - -## 激活时机 -- 当比较两个目录结构在内容上的差异时。 -- 当为特定文件夹(如配置或源代码)创建安全基准时。 -- 当通过基准哈希文件在版本控制中跟踪目录更改时。 - -## 核心原则与规范 -- **一致性**: 文件按字母顺序排序,以确保无论操作系统遍历顺序如何,都能获得一致的哈希输出。 -- **相对路径**: 在输出中使用相对路径,以便在不同的根路径之间实现可移植性。 - -## 实战示例 - -### 目录完整性 -```bash -# 为项目目录计算 SHA256 哈希列表 -x hashdir --sha256 ./my_project -``` - -### 基准比较 -```bash -# 将当前目录状态与保存的 MD5 基准进行比较 -x hashdir . | diff - baseline.md5 -``` - -## 交付验证清单 -- [ ] 确认目标目录路径。 -- [ ] 验证所需的哈希算法(默认为 MD5)。 -- [ ] 确保用户知晓该操作是递归的。 diff --git a/data/x-cmd/hi/SKILL.md b/data/x-cmd/hi/SKILL.md deleted file mode 100644 index ebd6654..0000000 --- a/data/x-cmd/hi/SKILL.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: hi -description: > - Display detailed HTTP information and network metadata. - Core Scenario: When the user needs a quick summary of their current HTTP environment or request headers. -license: MIT ---- - -# hi - HTTP Information Summary - -The `hi` module provides a quick way to view current HTTP metadata and request information, often used for debugging connectivity or viewing public IP info from an HTTP perspective. - -## When to Activate -- When the user wants to see their public-facing HTTP headers and IP. -- When performing basic network diagnostic checks related to web requests. - -## Patterns & Examples - -### View HTTP Info -```bash -# Display general HTTP environment and connection details -x hi -``` - -## Checklist -- [ ] Confirm if the user needs public IP info or header details. diff --git a/data/x-cmd/hi/SKILL.zh.md b/data/x-cmd/hi/SKILL.zh.md deleted file mode 100644 index ad78852..0000000 --- a/data/x-cmd/hi/SKILL.zh.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: hi -description: > - 显示详细的 HTTP 信息和网络元数据。 - 核心场景:当用户需要其当前 HTTP 环境或请求头的快速摘要时。 -license: MIT ---- - -# hi - HTTP 信息摘要 - -`hi` 模块提供了一种快速查看当前 HTTP 元数据和请求信息的方法,通常用于调试连通性或从 HTTP 角度查看公网 IP 信息。 - -## 激活时机 -- 当用户想要查看其面向公众的 HTTP 头和 IP 时。 -- 当执行与 Web 请求相关的基础网络诊断检查时。 - -## 实战示例 - -### 查看 HTTP 信息 -```bash -# 显示常规 HTTP 环境和连接详情 -x hi -``` - -## 交付验证清单 -- [ ] 确认用户需要的是公网 IP 信息还是请求头详情。 diff --git a/data/x-cmd/hn/SKILL.md b/data/x-cmd/hn/SKILL.md deleted file mode 100644 index fb2e390..0000000 --- a/data/x-cmd/hn/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: hn -description: > - Browse Hacker News from the CLI with interactive TUI, AI search integration, and user analysis. - Core Scenario: When the user wants to read top HN posts, search HN via AI, or analyze HN user stats. -license: MIT ---- - -# hn - Hacker News CLI Browser - -The `hn` module provides an interactive terminal interface for browsing Hacker News. It supports viewing top/new/best stories, AI-powered searching, and analyzing user metrics like h-index. - -## When to Activate -- When the user wants to browse Hacker News stories (top, new, ask, etc.) in an interactive table. -- When searching for specific topics on Hacker News using DuckDuckGo or AI (`::` prefix). -- When retrieving details for a specific post ID or user. -- When exporting HN data to JSON for scripting or analysis. - -## Core Principles & Rules -- **Interactive Browsing**: Optimized for terminal navigation with shortcuts to open links. -- **AI-Enhanced Search**: Use `ddgoai` or `::` to combine search results with AI summaries. -- **User Metrics**: Provides tools like `hidx` to calculate user impact. - -## Patterns & Examples - -### Browse Top Stories -```bash -# Open interactive TUI for top HN stories -x hn -``` - -### AI Search HN -```bash -# Search for 'llama3' on HN with AI-generated summaries -x hn :: llama3 -``` - -### View User Info -```bash -# Get details and h-index for a specific HN user -x hn user dang -x hn hi dang -``` - -## Checklist -- [ ] Confirm if the user wants to browse categories or perform a search. -- [ ] Verify if AI-enhanced search is preferred for complex queries. diff --git a/data/x-cmd/hn/SKILL.zh.md b/data/x-cmd/hn/SKILL.zh.md deleted file mode 100644 index ee04cf9..0000000 --- a/data/x-cmd/hn/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: hn -description: > - 通过 CLI 浏览 Hacker News,具有交互式 TUI、AI 搜索集成和用户分析功能。 - 核心场景:当用户想要阅读 HN 热门帖子、通过 AI 搜索 HN 或分析 HN 用户统计数据时。 -license: MIT ---- - -# hn - Hacker News 命令行浏览器 - -`hn` 模块提供了一个交互式终端界面来浏览 Hacker News。它支持查看热门/最新/最佳故事、AI 驱动的搜索,以及分析诸如 h-index 等用户指标。 - -## 激活时机 -- 当用户想要在交互式表格中浏览 Hacker News 故事(热门、最新、问答等)时。 -- 当使用 DuckDuckGo 或 AI(使用 `::` 前缀)在 Hacker News 上搜索特定主题时。 -- 当检索特定帖子 ID 或用户的详细信息时。 -- 当为脚本编写或分析导出 HN 数据到 JSON 时。 - -## 核心原则与规范 -- **交互式浏览**: 针对终端导航进行了优化,具有打开链接的快捷方式。 -- **AI 增强搜索**: 使用 `ddgoai` 或 `::` 将搜索结果与 AI 摘要相结合。 -- **用户指标**: 提供如 `hidx` 之类的工具来计算用户影响力。 - -## 实战示例 - -### 浏览热门故事 -```bash -# 打开热门 HN 故事的交互式 TUI -x hn -``` - -### AI 搜索 HN -```bash -# 在 HN 上搜索 'llama3' 并生成 AI 摘要 -x hn :: llama3 -``` - -### 查看用户信息 -```bash -# 获取特定 HN 用户的详情和 h-index -x hn user dang -x hn hi dang -``` - -## 交付验证清单 -- [ ] 确认用户是想要浏览分类还是执行搜索。 -- [ ] 验证对于复杂查询是否首选 AI 增强搜索。 diff --git a/data/x-cmd/host/SKILL.md b/data/x-cmd/host/SKILL.md deleted file mode 100644 index 58e911f..0000000 --- a/data/x-cmd/host/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: host -description: > - Securely manage /etc/hosts with interactive search, subcommands for adding/removing entries, and automated backups. - Core Scenario: When the user needs to edit local hostname-to-IP mappings or perform fuzzy searches in their hosts file. -license: MIT ---- - -# host - Local Host Table Management - -The `host` module provides a safe and efficient way to manage the local `/etc/hosts` file. It handles sudo permissions, provides automated backups before edits, and supports interactive fuzzy searching. - -## When to Activate -- When the user wants to add or remove local hostname-to-IP mappings. -- When performing a quick lookup of an IP for a local hostname. -- When searching the hosts file interactively using FZF (`fz`). -- When viewing the full contents of the hosts file with pagination. - -## Core Principles & Rules -- **Safety**: Automatically creates backups before any modification (`ed` subcommand). -- **Interactive Search**: Use `fz` for a searchable TUI experience. -- **Privilege Handling**: Clearly indicates that editing requires `sudo` access. - -## Patterns & Examples - -### Add/Edit Mapping -```bash -# Map a local development domain to an IP -x host ed myapp.local=192.168.1.100 -``` - -### Remove Mapping -```bash -# Delete a specific hostname mapping (prefix with -) -x host ed -myapp.local -``` - -### Search Hosts -```bash -# Interactively search for an entry in the hosts file -x host fz -``` - -## Checklist -- [ ] Confirm the target hostname and IP address for edits. -- [ ] Verify if the user has sudo permissions for modification. -- [ ] Ensure the operation (add/remove) is clearly understood. diff --git a/data/x-cmd/host/SKILL.zh.md b/data/x-cmd/host/SKILL.zh.md deleted file mode 100644 index 60a4281..0000000 --- a/data/x-cmd/host/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: host -description: > - 安全地管理 /etc/hosts,具有交互式搜索、添加/删除条目的子命令以及自动备份。 - 核心场景:当用户需要编辑本地主机名到 IP 的映射或在其 hosts 文件中执行模糊搜索时。 -license: MIT ---- - -# host - 本地主机表管理 - -`host` 模块提供了一种安全且高效的方式来管理本地的 `/etc/hosts` 文件。它处理 sudo 权限,在编辑前提供自动备份,并支持交互式模糊搜索。 - -## 激活时机 -- 当用户想要添加或删除本地主机名到 IP 的映射时。 -- 当执行本地主机名 IP 的快速查找时。 -- 当使用 FZF (`fz`) 交互式搜索 hosts 文件时。 -- 当通过分页查看 hosts 文件的完整内容时。 - -## 核心原则与规范 -- **安全性**: 在执行任何修改 (`ed` 子命令) 前自动创建备份。 -- **交互式搜索**: 使用 `fz` 获得可搜索的 TUI 体验。 -- **权限处理**: 明确指出编辑需要 `sudo` 访问权限。 - -## 实战示例 - -### 添加/编辑映射 -```bash -# 将本地开发域名映射到 IP -x host ed myapp.local=192.168.1.100 -``` - -### 删除映射 -```bash -# 删除特定的主机名映射(前缀加 -) -x host ed -myapp.local -``` - -### 搜索 Hosts -```bash -# 交互式搜索 hosts 文件中的条目 -x host fz -``` - -## 交付验证清单 -- [ ] 确认编辑的目标主机名和 IP 地址。 -- [ ] 验证用户是否具有修改所需的 sudo 权限。 -- [ ] 确保操作(添加/删除)被清晰理解。 diff --git a/data/x-cmd/hostname/SKILL.md b/data/x-cmd/hostname/SKILL.md deleted file mode 100644 index fb2f7fd..0000000 --- a/data/x-cmd/hostname/SKILL.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: hostname -description: > - Manage the system's hostname, including viewing current name and setting new names. - Core Scenario: When the user needs to identify the device name or change the system hostname. -license: MIT ---- - -# hostname - System Hostname Management - -The `hostname` module provides straightforward subcommands to retrieve or update the system's hostname, supporting both long and short formats. - -## When to Activate -- When identifying the system name for scripts or documentation. -- When changing the system's identity (requires elevated privileges). - -## Core Principles & Rules -- **Formats**: Support for `--long` (FQDN) and `--short` formats. -- **Permissions**: Changing the hostname with `set` requires `sudo` privileges. - -## Patterns & Examples - -### View Hostname -```bash -# Display the full system hostname -x hostname -# Display only the short hostname -x hostname --short -``` - -### Set Hostname -```bash -# Change the system hostname to a new value -x hostname set my-new-server -``` - -## Checklist -- [ ] Confirm if the long or short format is needed. -- [ ] Ensure the user has the required permissions if setting a new name. diff --git a/data/x-cmd/hostname/SKILL.zh.md b/data/x-cmd/hostname/SKILL.zh.md deleted file mode 100644 index b0a819f..0000000 --- a/data/x-cmd/hostname/SKILL.zh.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: hostname -description: > - 管理系统主机名,包括查看当前名称和设置新名称。 - 核心场景:当用户需要识别设备名称或更改系统主机名时。 -license: MIT ---- - -# hostname - 系统主机名管理 - -`hostname` 模块提供了直接的子命令来检索或更新系统的主机名,支持长格式和短格式。 - -## 激活时机 -- 当为脚本或文档识别系统名称时。 -- 当更改系统标识时(需要提升权限)。 - -## 核心原则与规范 -- **格式**: 支持 `--long` (FQDN) 和 `--short` 格式。 -- **权限**: 使用 `set` 更改主机名需要 `sudo` 权限。 - -## 实战示例 - -### 查看主机名 -```bash -# 显示完整系统主机名 -x hostname -# 仅显示短主机名 -x hostname --short -``` - -### 设置主机名 -```bash -# 将系统主机名更改为新值 -x hostname set my-new-server -``` - -## 交付验证清单 -- [ ] 确认需要长格式还是短格式。 -- [ ] 如果设置新名称,确保用户具有必要的权限。 diff --git a/data/x-cmd/hub/SKILL.md b/data/x-cmd/hub/SKILL.md deleted file mode 100644 index 4fe126b..0000000 --- a/data/x-cmd/hub/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: hub -description: > - Securely upload, host, and share scripts via x-cmd cloud, supporting AI-driven knowledge base search. - Core Scenario: When the user needs to manage cloud-hosted scripts or search the x-cmd knowledge base using AI. -license: MIT ---- - -# hub - Cloud Script Management & AI Search - -The `hub` module provides a secure space for managing and sharing scripts in the x-cmd cloud. It supports encrypted hosting, cross-device access, and an AI-powered search for the x-cmd knowledge base. - -## When to Activate -- When the user wants to upload a local script to the cloud for sharing or cross-device use. -- When performing an AI search within the x-cmd knowledge base for specific module usage. -- When compiling Go, Gop, or Zig scripts in a remote cloud environment. -- When managing x-cmd hub account and file permissions. - -## Core Principles & Rules -- **Security**: Supports encrypted hosting to protect script privacy. -- **Knowledge Retrieval**: Use the `search` or `ai` subcommands to leverage the AI knowledge base. -- **Access Control**: Use `access` to toggle between public and private file states. - -## Patterns & Examples - -### Upload and Share -```bash -# Upload a file and set it to public -x hub file put --public my_script.sh -``` - -### AI Search -```bash -# Search for how to add members in a GitHub repo via x-cmd -x hub search "how to add members in github repo" -``` - -### Account Info -```bash -# Check current cloud space usage and account details -x hub info -``` - -## Checklist -- [ ] Ensure the user is logged in via `x hub login`. -- [ ] Confirm if the file should be public or private. -- [ ] Verify if the AI search query is specific enough. diff --git a/data/x-cmd/hub/SKILL.zh.md b/data/x-cmd/hub/SKILL.zh.md deleted file mode 100644 index cc6449b..0000000 --- a/data/x-cmd/hub/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: hub -description: > - 通过 x-cmd 云安全地上传、托管和共享脚本,支持 AI 驱动的知识库搜索。 - 核心场景:当用户需要管理云端托管脚本或使用 AI 搜索 x-cmd 知识库时。 -license: MIT ---- - -# hub - 云端脚本管理与 AI 搜索 - -`hub` 模块提供了一个安全的空间,用于在 x-cmd 云中管理和共享脚本。它支持加密托管、跨设备访问以及针对 x-cmd 知识库的 AI 驱动搜索。 - -## 激活时机 -- 当用户想要将本地脚本上传到云端以便共享或跨设备使用时。 -- 当在 x-cmd 知识库中执行 AI 搜索以查询特定模块的用法时。 -- 当在远程云环境中编译 Go、Gop 或 Zig 脚本时。 -- 当管理 x-cmd hub 账户和文件权限时。 - -## 核心原则与规范 -- **安全性**: 支持加密托管以保护脚本隐私。 -- **知识检索**: 使用 `search` 或 `ai` 子命令来利用 AI 知识库。 -- **访问控制**: 使用 `access` 在公开和私有文件状态之间进行切换。 - -## 实战示例 - -### 上传并共享 -```bash -# 上传文件并将其设置为公开 -x hub file put --public my_script.sh -``` - -### AI 搜索 -```bash -# 搜索如何通过 x-cmd 在 GitHub 仓库中添加成员 -x hub search "如何在 github 仓库中添加成员" -``` - -### 账户信息 -```bash -# 检查当前的云端空间使用情况和账户详情 -x hub info -``` - -## 交付验证清单 -- [ ] 确保用户已通过 `x hub login` 登录。 -- [ ] 确认文件应该是公开的还是私有的。 -- [ ] 验证 AI 搜索查询是否足够具体。 diff --git a/data/x-cmd/humantime/SKILL.md b/data/x-cmd/humantime/SKILL.md deleted file mode 100644 index f4fff58..0000000 --- a/data/x-cmd/humantime/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: humantime -description: > - Convert between seconds and human-readable time formats (e.g., 1h30m). - Core Scenario: When the user needs to parse or display durations in scripts or terminal output. -license: MIT ---- - -# humantime - Human-Friendly Time Conversion - -The `humantime` module simplifies time manipulation by converting raw seconds into readable formats (like `1h 30m`) and vice-versa. It supports fractional seconds and automatic type detection. - -## When to Activate -- When displaying script execution duration to users. -- When calculating cache expiration times based on human-readable input. -- When checking if a duration falls within a specific limit (`in` subcommand). - -## Core Principles & Rules -- **Auto-Detection**: Inputting an integer (e.g., `3600`) converts to time format (`1h`), while a time string (e.g., `5m`) converts to seconds (`300`). -- **Time Units**: Supports `d` (days), `h` (hours), `m` (minutes), and `s` (seconds). -- **Fractional Precision**: Handles decimal seconds (e.g., `3661.5s = 1h1m1.5s`) accurately. - -## Patterns & Examples - -### Convert Seconds to Readable -```bash -# Convert one day into readable format -x humantime 86400 -``` - -### Convert Readable to Seconds -```bash -# Get raw seconds for 2 hours and 30 minutes -x humantime 2h30m -``` - -### Within Range Check -```bash -# Check if 240 seconds is within 5 minutes (returns 0 for yes) -x humantime in 5m 240 -``` - -## Checklist -- [ ] Confirm if the input is an integer (seconds) or a string (readable). -- [ ] Verify if high precision (fractional seconds) is required. diff --git a/data/x-cmd/humantime/SKILL.zh.md b/data/x-cmd/humantime/SKILL.zh.md deleted file mode 100644 index 9b61612..0000000 --- a/data/x-cmd/humantime/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: humantime -description: > - 在秒数和人性化时间格式(如 1h30m)之间进行转换。 - 核心场景:当用户需要在脚本或终端输出中解析或显示时长时。 -license: MIT ---- - -# humantime - 人性化时间转换 - -`humantime` 模块通过将原始秒数转换为可读格式(如 `1h 30m`)以及反向转换,简化了时间操作。它支持小数秒和自动类型检测。 - -## 激活时机 -- 当向用户显示脚本执行时长时。 -- 当根据人性化输入计算缓存过期时间时。 -- 当检查一个时长是否在特定限制内时(使用 `in` 子命令)。 - -## 核心原则与规范 -- **自动检测**: 输入整数(例如 `3600`)会转换为时间格式 (`1h`),而输入时间字符串(例如 `5m`)会转换为秒数 (`300`)。 -- **时间单位**: 支持 `d` (天)、`h` (小时)、`m` (分钟) 和 `s` (秒)。 -- **高精度**: 能够准确处理小数秒(例如 `3661.5s = 1h1m1.5s`)。 - -## 实战示例 - -### 将秒数转换为可读格式 -```bash -# 将一天的时间转换为可读格式 -x humantime 86400 -``` - -### 将可读格式转换为秒数 -```bash -# 获取 2 小时 30 分钟的原始秒数 -x humantime 2h30m -``` - -### 范围内检查 -```bash -# 检查 240 秒是否在 5 分钟内(如果是,则返回 0) -x humantime in 5m 240 -``` - -## 交付验证清单 -- [ ] 确认输入是整数(秒)还是字符串(可读格式)。 -- [ ] 验证是否需要高精度(小数秒)。 diff --git a/data/x-cmd/id/SKILL.md b/data/x-cmd/id/SKILL.md deleted file mode 100644 index 2b6671f..0000000 --- a/data/x-cmd/id/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: id -description: > - Enhanced user identity information with colorful and structured output. - Core Scenario: When the user needs to check UID, GID, and group memberships in a highly readable format. -license: MIT ---- - -# id - Enhanced User Identity - -The `id` module provides a more readable and colorful alternative to the standard `id` command, showing user and group IDs along with memberships in a structured key-value format. - -## When to Activate -- When the user wants to view their own or another user's identity info (UID, GID). -- When a script needs to retrieve just the numeric UID or GID. -- When performing manual verification of group memberships. - -## Core Principles & Rules -- **Readability**: Automatically provides colorful output in TTY sessions. -- **Script-Friendly**: Use `uid_` or `gid_` subcommands to store results directly in variables for automation. -- **Native Fallback**: Reverts to standard behavior for unknown options. - -## Patterns & Examples - -### View Identity -```bash -# Show colorful identity info for the current user -x id -``` - -### Check Specific User -```bash -# View identity details for 'root' -x id root -``` - -### Get Numeric UID -```bash -# Output only the numeric User ID -x id uid -``` - -## Checklist -- [ ] Confirm if the user is checking their own ID or another user's. -- [ ] Verify if only the numeric ID is required for a script. diff --git a/data/x-cmd/id/SKILL.zh.md b/data/x-cmd/id/SKILL.zh.md deleted file mode 100644 index d2fbd05..0000000 --- a/data/x-cmd/id/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: id -description: > - 具有彩色和结构化输出的增强型用户身份信息。 - 核心场景:当用户需要以高可读性格式检查 UID、GID 和组关系时。 -license: MIT ---- - -# id - 增强型用户身份查看 - -`id` 模块为标准 `id` 命令提供了一个更具可读性且彩色的替代方案,以结构化的键值格式显示用户和组 ID 及其成员关系。 - -## 激活时机 -- 当用户想要查看自己或其他用户的身份信息 (UID, GID) 时。 -- 当脚本需要仅检索数字 UID 或 GID 时。 -- 当手动验证组关系时。 - -## 核心原则与规范 -- **可读性**: 在 TTY 会话中自动提供彩色输出。 -- **脚本友好**: 使用 `uid_` 或 `gid_` 子命令将结果直接存储在变量中以进行自动化。 -- **原生回退**: 对于未知选项,回退到标准行为。 - -## 实战示例 - -### 查看身份 -```bash -# 为当前用户显示彩色的身份信息 -x id -``` - -### 检查特定用户 -```bash -# 查看 'root' 的身份详情 -x id root -``` - -### 获取数字 UID -```bash -# 仅输出数字形式的用户 ID -x id uid -``` - -## 交付验证清单 -- [ ] 确认用户是在检查自己的 ID 还是其他用户的。 -- [ ] 验证脚本是否仅需要数字 ID。 diff --git a/data/x-cmd/ip/SKILL.md b/data/x-cmd/ip/SKILL.md deleted file mode 100644 index 7f97930..0000000 --- a/data/x-cmd/ip/SKILL.md +++ /dev/null @@ -1,53 +0,0 @@ ---- -name: ip -description: > - Network diagnostics and IP geolocation query. - Core Scenario: When AI needs to query local network configuration or geographic info of remote server IPs. -license: MIT ---- - -# x ip - Network Diagnostics & Geolocation (AI Optimized) - -The `x ip` module provides a concise interface for querying IP-related information, covering both local NIC configurations and global IP geographic data. - -## When to Activate -- When retrieving the local public IP or internal network interface addresses. -- When querying geographic location, ISP, or other info for a specific IP. -- When scanning active hosts or open ports in a subnet (network diagnostics). - -## Core Principles & Rules -- **Non-interactive First**: Avoid interactive configurations; retrieve data directly via subcommands. -- **Structured Thinking**: Use subcommands like `geolite` for detailed geographic insights. - -## Patterns & Examples - -### Query Local Public IP and Location -```bash -# Get detailed geographic info for the current public IP -x ip geolite -``` - -### Query Geolocation for a Specific IP -```bash -# Query detailed location info for 8.8.8.8 -x ip geolite 8.8.8.8 -``` - -### List All Local Interface Addresses -```bash -# Get a list of all IP addresses on the machine (non-interactive) -x ip ls -``` - -### Network Scanning (Diagnostics) -```bash -# Scan active hosts in a subnet (may require privileges) -x ip map 192.168.1.0/24 - -# Quickly scan common ports for a specific IP -x ip tps 127.0.0.1 -``` - -## Checklist -- [ ] Confirm whether querying a specific IP or the local egress IP. -- [ ] Check privilege requirements before running scan tasks. diff --git a/data/x-cmd/ip/SKILL.zh.md b/data/x-cmd/ip/SKILL.zh.md deleted file mode 100644 index a895e57..0000000 --- a/data/x-cmd/ip/SKILL.zh.md +++ /dev/null @@ -1,53 +0,0 @@ ---- -name: ip -description: > - 网络诊断与 IP 地理位置查询。 - 核心场景:当 AI 需要查询本机网络配置或远程服务器 IP 的地理信息时。 -license: MIT ---- - -# x ip - 网络诊断与地理查询 (AI 优化版) - -`x ip` 模块提供了简洁的接口来查询 IP 地址相关信息,无论是本机网卡配置还是全球 IP 的地理归属。 - -## 激活时机 -- 当需要获取本机公网 IP 或内网网卡地址时。 -- 当需要查询特定 IP 地址的地理位置、所属 ISP 等信息时。 -- 当需要扫描子网中的活跃主机或开放端口时(网络诊断)。 - -## 核心原则与规范 -- **非交互优先**: 避免使用交互式配置,直接通过子命令获取数据。 -- **结构化思维**: 部分子命令(如 `geolite`)提供详细的地理信息。 - -## 实战示例 - -### 查询本机公网 IP 及地理位置 -```bash -# 获取当前公网出口 IP 的详细地理信息 -x ip geolite -``` - -### 查询特定 IP 的归属地 -```bash -# 查询 8.8.8.8 的详细位置信息 -x ip geolite 8.8.8.8 -``` - -### 列出所有本地网卡地址 -```bash -# 获取本机所有 IP 地址列表(非交互) -x ip ls -``` - -### 网络扫描 (诊断任务) -```bash -# 扫描子网内的活跃主机 (需要权限) -x ip map 192.168.1.0/24 - -# 快速扫描特定 IP 的常见端口 -x ip tps 127.0.0.1 -``` - -## 交付验证清单 -- [ ] 确认是否需要查询特定 IP 还是本机出口 IP。 -- [ ] 在执行扫描任务前确认权限要求。 diff --git a/data/x-cmd/java/SKILL.md b/data/x-cmd/java/SKILL.md deleted file mode 100644 index d5db8b3..0000000 --- a/data/x-cmd/java/SKILL.md +++ /dev/null @@ -1,28 +0,0 @@ ---- -name: java -description: > - Enhanced Java (JDK) environment manager for installing, switching, and running Java applications. - Core Scenario: When the user needs to manage JDK versions or run Java/Jar files via CLI. -license: MIT ---- - -# java - Java (JDK) Environment Management - -The `java` module provides a comprehensive way to manage Java Development Kits (JDK) and execute Java applications, ensuring that the appropriate runtime is available across different systems. - -## When to Activate -- When installing or switching between multiple JDK versions. -- When running `.jar` files or compiling Java source code. -- When identifying the currently active Java environment. - -## Patterns & Examples - -### Run Jar File -```bash -# Execute a Java archive file -x java -jar myapp.jar -``` - -## Checklist -- [ ] Verify the desired JDK version. -- [ ] Ensure the `.jar` or `.java` file path is correct. diff --git a/data/x-cmd/java/SKILL.zh.md b/data/x-cmd/java/SKILL.zh.md deleted file mode 100644 index 0c316d4..0000000 --- a/data/x-cmd/java/SKILL.zh.md +++ /dev/null @@ -1,28 +0,0 @@ ---- -name: java -description: > - 增强型 Java (JDK) 环境管理器,用于安装、切换和运行 Java 应用程序。 - 核心场景:当用户需要管理 JDK 版本或通过 CLI 运行 Java/Jar 文件时。 -license: MIT ---- - -# java - Java (JDK) 环境管理 - -`java` 模块提供了一套全面的管理 Java 开发工具包 (JDK) 和执行 Java 应用程序的方法,确保在不同系统上都有合适的运行时可用。 - -## 激活时机 -- 当安装或在多个 JDK 版本之间切换时。 -- 当运行 `.jar` 文件或编译 Java 源代码时。 -- 当识别当前激活的 Java 环境时。 - -## 实战示例 - -### 运行 Jar 文件 -```bash -# 执行一个 Java 归档文件 -x java -jar myapp.jar -``` - -## 交付验证清单 -- [ ] 验证所需的 JDK 版本。 -- [ ] 确保 `.jar` 或 `.java` 文件路径正确。 diff --git a/data/x-cmd/jina/SKILL.md b/data/x-cmd/jina/SKILL.md deleted file mode 100644 index b488d09..0000000 --- a/data/x-cmd/jina/SKILL.md +++ /dev/null @@ -1,58 +0,0 @@ ---- -name: jina -description: > - Jina AI enhancement tool: Web parsing, search, and reranking. - Core Scenario: When AI needs to read a URL's content, perform real-time web searches, or rank text segments by semantic relevance. -license: MIT ---- - -# x jina - AI Knowledge Assistant (AI Optimized) - -`x jina` serves as the "eyes" and "external brain" for AI Agents. Its core strength lies in converting complex web pages into AI-friendly Markdown and providing powerful search and semantic ranking capabilities. - -## When to Activate -- When reading and analyzing the content of a specific web page (URL), automatically converted to Markdown. -- When performing real-time web searches to gather the latest information. -- When identifying the most semantically relevant segments from a set of documents (Reranker). - -## Core Principles & Rules -- **Markdown First**: Retrieves web pages in Markdown by default, which is optimal for LLM consumption. -- **Environment Requirements**: Requires a Jina API Key. If not configured, AI should guide the user through initialization. -- **Configuration Guidance**: - - Direct the user to jina.ai to obtain a free or paid API Key. - - Suggest the user run `x jina init` for configuration. - -## Patterns & Examples - -### Read Web Content (Most Common) -```bash -# Retrieve web content and convert to Markdown for AI reading -x jina reader https://example.com/article -``` - -### Real-time Web Search -```bash -# Search for keywords and return summaries of the top 5 results -x jina search "latest AI trends 2024" -``` - -### Semantic Reranking -```bash -# Find the top 3 segments from a file most relevant to "how to install" -x jina reranker generate -f docs.txt --top 3 "how to install" -``` - -### Text Embedding -```bash -# Generate vector data for text, useful for vector database retrieval -x jina embed generate "This is a test text" -``` - -## Configuration Guide (for AI) -If an API error occurs, provide this guidance to the user: -> Please obtain an API Key from Jina AI (https://jina.ai), then run the following command in your terminal to initialize: -> `x jina init` - -## Checklist -- [ ] Prioritize using the `reader` subcommand for Markdown. -- [ ] Combine search tasks with `reranker` for better precision. diff --git a/data/x-cmd/jina/SKILL.zh.md b/data/x-cmd/jina/SKILL.zh.md deleted file mode 100644 index f4193a1..0000000 --- a/data/x-cmd/jina/SKILL.zh.md +++ /dev/null @@ -1,58 +0,0 @@ ---- -name: jina -description: > - Jina AI 增强工具:网页解析、联网搜索与语义重排。 - 核心场景:当 AI 需要阅读某个 URL 的内容、进行实时联网搜索或对大量文本进行语义相关性排序时。 -license: MIT ---- - -# x jina - AI 知识增强助手 (AI 优化版) - -`x jina` 是 AI Agent 的“眼睛”和“外脑”,核心优势在于将复杂的网页转换为 AI 易读的 Markdown,并提供强大的联网搜索与语义筛选能力。 - -## 激活时机 -- 当需要阅读并分析某个网页 (URL) 的具体内容时(自动转为 Markdown)。 -- 当需要进行实时联网搜索以获取最新资讯时。 -- 当需要从一组文档中找出与当前问题语义最相关的片段时(Reranker)。 - -## 核心原则与规范 -- **Markdown 优先**: 默认以 Markdown 格式获取网页,这最适合 LLM 阅读。 -- **环境要求**: 需要 Jina API Key。如果未配置,AI 应引导用户初始化。 -- **配置引导**: - - 引导用户访问 jina.ai 获取免费或付费的 API Key。 - - 建议用户运行 `x jina init` 进行配置。 - -## 实战示例 - -### 阅读网页内容 (最常用) -```bash -# 获取网页内容并转换为 Markdown (AI 直接阅读) -x jina reader https://example.com/article -``` - -### 联网搜索 -```bash -# 搜索关键词并返回前 5 条结果的摘要 -x jina search "最新 AI 趋势 2024" -``` - -### 语义相关性重排 (Reranker) -```bash -# 从文件中找出与 "如何安装" 最相关的 3 个片段 -x jina reranker generate -f docs.txt --top 3 "如何安装" -``` - -### 生成文本向量 (Embedding) -```bash -# 为文本生成向量数据,用于后续向量数据库检索 -x jina embed generate "这是一段测试文本" -``` - -## 配置指南 (针对 AI) -如果遇到 API 错误,请向用户输出以下引导: -> 请先在 Jina AI 官网 (https://jina.ai) 获取 API Key,然后在终端运行以下命令进行初始化: -> `x jina init` - -## 交付验证清单 -- [ ] 优先使用 `reader` 子命令获取 Markdown。 -- [ ] 搜索任务建议结合 `reranker` 提升精准度。 diff --git a/data/x-cmd/jq/SKILL.md b/data/x-cmd/jq/SKILL.md deleted file mode 100644 index 183eb75..0000000 --- a/data/x-cmd/jq/SKILL.md +++ /dev/null @@ -1,56 +0,0 @@ ---- -name: jq -description: > - Lightweight and flexible JSON processor. - Core Scenario: When the AI needs to filter, transform, format, or extract JSON data in the terminal. x-cmd version provides zero-dependency auto-installation. -license: MIT ---- - -# x jq - JSON Processor (AI Optimized) - -`x jq` is an enhanced module based on jq, with the core advantage of **zero-dependency auto-installation** and **optimization for scripted tasks**. It ensures JSON can be processed immediately in any environment. - -## When to Activate -- When specific fields need to be extracted from complex JSON responses. -- When JSON structures need to be filtered, restructured, or transformed. -- When unformatted JSON strings need to be beautified for further analysis. -- When processing results need to be output as raw strings (`-r`) for other commands. - -## Core Principles & Rules -- **Non-interactive First**: AI should avoid the interactive `repl` mode and use jq expressions directly. -- **Pipe Integration**: Recommended for use with pipes, e.g., `cat data.json | x jq '.field'`. -- **Format Control**: - - Use `-r` (raw-output) for raw values without quotes (ideal for getting single string values). - - Use `-c` (compact-output) for compact one-line JSON to save context Tokens. -- **Environment Isolation**: `x jq` automatically downloads and runs jq when necessary, without polluting the system environment. - -## Patterns & Examples - -### Extract and Output Raw String -```bash -# Get the value of the 'version' field (no quotes, suitable for script use) -x jq -r '.version' package.json -``` - -### Filter Array and Compact Output -```bash -# Filter and output as a compact one-line JSON to save Tokens -x jq -c '.items[] | select(.status == "active")' data.json -``` - -### Construct New JSON Object -```bash -# Construct a new object containing status and timestamp -x jq -n --arg ts "$(date)" '{"status": "ok", "timestamp": $ts}' -``` - -### Process Multiple Files -```bash -# Merge and process multiple JSON files -x jq -s '.[0] * .[1]' config1.json config2.json -``` - -## Checklist -- [ ] Confirm that a non-interactive command is used (no `r` or `repl`). -- [ ] Consider if `-r` is needed for plain text values. -- [ ] Consider if `-c` is needed to reduce output volume. diff --git a/data/x-cmd/jq/SKILL.zh.md b/data/x-cmd/jq/SKILL.zh.md deleted file mode 100644 index 6c86a59..0000000 --- a/data/x-cmd/jq/SKILL.zh.md +++ /dev/null @@ -1,56 +0,0 @@ ---- -name: jq -description: > - 轻量级且灵活的 JSON 处理器。 - 核心场景:当 AI 需要在终端中过滤、转换、格式化或提取 JSON 数据时。x-cmd 版 jq 提供零依赖自动安装。 -license: MIT ---- - -# x jq - JSON 处理器 (AI 优化版) - -`x jq` 是基于 jq 的增强模块,核心优势在于 **零依赖自动安装** 和 **针对脚本化任务的优化**。它能确保在任何环境下都能立即处理 JSON。 - -## 激活时机 -- 当需要从复杂的 JSON 响应中提取特定字段时。 -- 当需要过滤、重构或转换 JSON 结构时。 -- 当需要将非格式化的 JSON 字符串美化以供后续分析时。 -- 当需要将处理结果以原始字符串格式(`-r`)输出给其他命令时。 - -## 核心原则与规范 -- **非交互优先**: AI 应当避免使用交互式 `repl` 模式,应直接使用 jq 表达式。 -- **管道集成**: 建议配合管道使用,如 `cat data.json | x jq '.field'`。 -- **格式控制**: - - 使用 `-r` (raw-output) 输出不带引号的原始值(适合获取单个字符串值)。 - - 使用 `-c` (compact-output) 输出紧凑的一行 JSON,节省上下文 Token。 -- **环境隔离**: `x jq` 会在必要时自动下载并运行 jq,不会污染系统环境。 - -## 实战示例 - -### 提取并输出原始字符串 -```bash -# 获取 'version' 字段的值(不带引号,适合后续脚本使用) -x jq -r '.version' package.json -``` - -### 过滤数组并紧凑输出 -```bash -# 过滤并以一行紧凑 JSON 输出,节省 Token -x jq -c '.items[] | select(.status == "active")' data.json -``` - -### 构造新的 JSON 对象 -```bash -# 构造一个包含 status 和 timestamp 的新对象 -x jq -n --arg ts "$(date)" '{"status": "ok", "timestamp": $ts}' -``` - -### 处理多个文件 -```bash -# 合并并处理多个 JSON 文件 -x jq -s '.[0] * .[1]' config1.json config2.json -``` - -## 交付验证清单 -- [ ] 确认使用的是非交互式命令(不带 `r` 或 `repl`)。 -- [ ] 考虑是否需要使用 `-r` 来获取纯文本值。 -- [ ] 考虑是否需要 `-c` 来减小输出体积。 diff --git a/data/x-cmd/kev/SKILL.md b/data/x-cmd/kev/SKILL.md deleted file mode 100644 index cf9988e..0000000 --- a/data/x-cmd/kev/SKILL.md +++ /dev/null @@ -1,36 +0,0 @@ ---- -name: kev -description: > - Retrieve and list Known Exploited Vulnerabilities (KEV) from official security catalogs. - Core Scenario: When the user needs to audit system safety against actively exploited security flaws. -license: MIT ---- - -# kev - Known Exploited Vulnerabilities Catalog - -The `kev` module provides a simple interface to list and query known exploited vulnerabilities, helping security professionals prioritize patching based on real-world exploit data. - -## When to Activate -- When performing a rapid audit of current security threats. -- When listing the top or recent vulnerabilities that are known to be active in the wild. - -## Core Principles & Rules -- **Direct Retrieval**: Fetches data from authoritative security catalogs. -- **Filtering**: Supports viewing the full list or just the top N most critical/recent entries. - -## Patterns & Examples - -### List All KEVs -```bash -# Display the entire catalog of known exploited vulnerabilities -x kev ls -``` - -### Top Critical KEVs -```bash -# View the top 100 entries from the KEV list -x kev top 100 -``` - -## Checklist -- [ ] Confirm if the user needs the full list or a limited summary. diff --git a/data/x-cmd/kev/SKILL.zh.md b/data/x-cmd/kev/SKILL.zh.md deleted file mode 100644 index a5af4fe..0000000 --- a/data/x-cmd/kev/SKILL.zh.md +++ /dev/null @@ -1,36 +0,0 @@ ---- -name: kev -description: > - 从官方安全目录中检索并罗列已知被利用的漏洞 (KEV)。 - 核心场景:当用户需要审计系统安全性,防范正被积极利用的安全缺陷时。 -license: MIT ---- - -# kev - 已知被利用漏洞清单 - -`kev` 模块提供了一个简单的界面来列出和查询已知被利用的漏洞,帮助安全专业人员根据现实世界的利用数据优先处理补丁。 - -## 激活时机 -- 当对当前安全威胁执行快速审计时。 -- 当列出已知在野外活跃的顶级或最新漏洞时。 - -## 核心原则与规范 -- **直接检索**: 从权威的安全目录中获取数据。 -- **过滤**: 支持查看完整列表或仅查看前 N 个最关键/最新的条目。 - -## 实战示例 - -### 列出所有 KEV -```bash -# 显示完整的已知被利用漏洞目录 -x kev ls -``` - -### 顶级关键 KEV -```bash -# 查看 KEV 清单中的前 100 条记录 -x kev top 100 -``` - -## 交付验证清单 -- [ ] 确认用户需要完整列表还是有限的摘要。 diff --git a/data/x-cmd/kill/SKILL.md b/data/x-cmd/kill/SKILL.md deleted file mode 100644 index fc7ba38..0000000 --- a/data/x-cmd/kill/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: kill -description: > - Enhanced kill command with support for recursive termination and killing processes by port. - Core Scenario: When the user needs to terminate groups of processes or free up specific network ports (80/443). -license: MIT ---- - -# kill - Enhanced Process Termination - -The `kill` module extends the system's ability to terminate processes by providing advanced features like killing process trees recursively or targeting processes based on the network ports they occupy. - -## When to Activate -- When the user needs to terminate all processes listening on standard ports (80, 443). -- When a TUI visualization of available signals is required. -- When performing recursive process termination. -- When checking signal lists in JSON format for automation. - -## Core Principles & Rules -- **Targeted Termination**: Use `byport` to specifically clear web service ports. -- **Visual Signal List**: Use `-l` for an interactive TUI to explore signals. -- **JSON Support**: Export signals to JSON for easy programmatic access. - -## Patterns & Examples - -### Kill by Port -```bash -# Terminate all processes listening on common web ports (currently supports 80 and 443) -x kill byport -``` - -### Signal List (TUI) -```bash -# Interactively browse available system signals -x kill -l -``` - -### Signal List (JSON) -```bash -# Get signal information in JSON format -x kill -l --json -``` - -## Checklist -- [ ] Confirm the specific PID or port to be targeted. -- [ ] Verify the signal type (default is SIGTERM) if manual signals are used. -- [ ] Ensure the user has the required permissions to kill the target processes. diff --git a/data/x-cmd/kill/SKILL.zh.md b/data/x-cmd/kill/SKILL.zh.md deleted file mode 100644 index 06f904c..0000000 --- a/data/x-cmd/kill/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: kill -description: > - 增强型 kill 命令,支持递归终止和按端口杀死进程。 - 核心场景:当用户需要终止进程组或释放特定的网络端口(80/443)时。 -license: MIT ---- - -# kill - 增强型进程终止工具 - -`kill` 模块通过提供高级功能(如递归终止进程树或根据进程占用的网络端口进行针对性终止),扩展了系统终止进程的能力。 - -## 激活时机 -- 当用户需要终止监听标准端口(80, 443)的所有进程时。 -- 当需要可用信号的 TUI 视觉呈现时。 -- 当执行递归进程终止时。 -- 当为了自动化而检查 JSON 格式的信号列表时。 - -## 核心原则与规范 -- **针对性终止**: 使用 `byport` 专门清理 Web 服务端口。 -- **视觉信号列表**: 使用 `-l` 通过交互式 TUI 探索信号。 -- **JSON 支持**: 将信号导出为 JSON 以便于编程访问。 - -## 实战示例 - -### 按端口杀死进程 -```bash -# 终止监听 80 和 443 端口的所有进程 -x kill byport -``` - -### 信号列表 (TUI) -```bash -# 交互式浏览可用的系统信号 -x kill -l -``` - -### 信号列表 (JSON) -```bash -# 以 JSON 格式获取信号信息 -x kill -l --json -``` - -## 交付验证清单 -- [ ] 确认要针对的具体 PID 或端口。 -- [ ] 如果使用手动信号,请验证信号类型(默认为 SIGTERM)。 -- [ ] 确保用户具有杀死目标进程所需的必要权限。 diff --git a/data/x-cmd/kimi/SKILL.md b/data/x-cmd/kimi/SKILL.md deleted file mode 100644 index 17c8d68..0000000 --- a/data/x-cmd/kimi/SKILL.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: kimi -description: > - Enhancement module for kimi-cli, integrating AI programming capabilities into the terminal workflow. - Core Scenario: When the user wants to launch the Kimi Code agent for coding assistance, upgrades, or session management. -license: MIT ---- - -# kimi - Kimi Code CLI Enhancement - -The `kimi` module enhances the `kimi-cli` agent, providing a seamless terminal experience for AI-driven coding, session management, and workspace integration. - -## When to Activate -- When the user wants to launch the Kimi Code interactive session. -- When the user needs to upgrade or manage the `kimi-cli` tool. -- When integrating Model Context Protocol (MCP) servers with Kimi. -- When continuing or forking previous AI coding sessions. - -## Core Principles & Rules -- **Yolo Mode**: Use the `-y` or `--yolo` flag for automatic tool approval if requested. -- **Thinking Mode**: Enable or disable deep thinking using `--thinking` or `--no-thinking`. -- **Environment Management**: Use `--install` or `--upgrade` to ensure the tool is up to date. - -## Additional Scenarios -- **TUI Mode**: Run the interactive terminal UI using `x kimi term`. -- **Web Interface**: Launch the Kimi web UI via `x kimi web`. - -## Patterns & Examples - -### Launch Kimi Code -```bash -# Start an interactive Kimi Code session in the current directory -x kimi -``` - -### Auto-Approve Commands -```bash -# Run Kimi with automatic approval for all actions (YOLO mode) -x kimi --yolo -``` - -### Continue Last Session -```bash -# Resume the most recent conversation in the current workspace -x kimi --continue -``` - -## Checklist -- [ ] Ensure `kimi-cli` is installed; run `x kimi --install` if necessary. -- [ ] Confirm if the user wants auto-approval (YOLO) enabled. -- [ ] Check if MCP configurations need loading. diff --git a/data/x-cmd/kimi/SKILL.zh.md b/data/x-cmd/kimi/SKILL.zh.md deleted file mode 100644 index 3f83119..0000000 --- a/data/x-cmd/kimi/SKILL.zh.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: kimi -description: > - kimi-cli 增强模块,将 AI 编程能力集成到终端工作流中。 - 核心场景:当用户想要启动 Kimi Code 代理进行编码辅助、升级工具或管理会话时。 -license: MIT ---- - -# kimi - Kimi Code CLI 增强 - -`kimi` 模块增强了 `kimi-cli` 代理,为 AI 驱动的编码、会话管理和工作区集成提供无缝的终端体验。 - -## 激活时机 -- 当用户想要启动 Kimi Code 交互式会话时。 -- 当用户需要升级或管理 `kimi-cli` 工具时。 -- 当将模型上下文协议 (MCP) 服务器与 Kimi 集成时。 -- 当继续或分叉之前的 AI 编码会话时。 - -## 核心原则与规范 -- **Yolo 模式**: 如果用户要求,使用 `-y` 或 `--yolo` 标志进行自动工具批准。 -- **思考模式**: 使用 `--thinking` 或 `--no-thinking` 启用或禁用深度思考。 -- **环境管理**: 使用 `--install` 或 `--upgrade` 确保工具是最新的。 - -## 补充场景 -- **TUI 模式**: 使用 `x kimi term` 运行交互式终端 UI。 -- **Web 界面**: 通过 `x kimi web` 启动 Kimi Web 界面。 - -## 实战示例 - -### 启动 Kimi Code -```bash -# 在当前目录下启动交互式 Kimi Code 会话 -x kimi -``` - -### 自动批准命令 -```bash -# 以自动批准所有操作的模式 (YOLO mode) 运行 Kimi -x kimi --yolo -``` - -### 继续上次会话 -```bash -# 恢复当前工作区中最近的对话 -x kimi --continue -``` - -## 交付验证清单 -- [ ] 确保已安装 `kimi-cli`;如有必要,运行 `x kimi --install`。 -- [ ] 确认用户是否想要启用自动批准 (YOLO)。 -- [ ] 检查是否需要加载 MCP 配置。 diff --git a/data/x-cmd/last/SKILL.md b/data/x-cmd/last/SKILL.md deleted file mode 100644 index e13d4d3..0000000 --- a/data/x-cmd/last/SKILL.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: last -description: > - Display user login history with interactive TUI, tree views, and structured data exports. - Core Scenario: When the user needs to audit login/logout records, reboots, or session durations. -license: MIT ---- - -# last - Enhanced Login History - -The `last` module provides a highly flexible way to view system login history, supporting interactive exploration, tree-based grouping by reboot, and structured data exports (JSON/CSV). - -## When to Activate -- When the user wants to audit recent system login and logout activity. -- When checking system reboot history in a structured format. -- When performing interactive analysis of user sessions in a TTY. -- When exporting login records for security reporting or external analysis. - -## Core Principles & Rules -- **Interactive Browsing**: Default behavior in TTY is an interactive selector for records. -- **Tree View**: Use `--tree` to group sessions by the reboot events that preceded them. -- **Structured Formats**: Support for `-j` (JSON) and `-c` (CSV) for easy parsing. - -## Patterns & Examples - -### Interactive Audit -```bash -# Browse all login/logout records interactively -x last -``` - -### Reboot History -```bash -# Show only system reboot records -x last --reboot -``` - -### JSON Export -```bash -# Get all login history as a JSON object -x last --json -``` - -### Grouped View -```bash -# Display login records grouped by system reboots in a tree structure -x last --tree -``` - -## Checklist -- [ ] Confirm if the user needs a specific format (JSON, CSV, Tree). -- [ ] Verify if filtering by user or reboot is required. diff --git a/data/x-cmd/last/SKILL.zh.md b/data/x-cmd/last/SKILL.zh.md deleted file mode 100644 index f7798a8..0000000 --- a/data/x-cmd/last/SKILL.zh.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: last -description: > - 通过交互式 TUI、树状视图和结构化数据导出显示用户登录历史。 - 核心场景:当用户需要审核登录/注销记录、重启历史或会话时长时。 -license: MIT ---- - -# last - 增强型登录历史查看 - -`last` 模块提供了一种高度灵活的方式来查看系统登录历史,支持交互式探索、按重启进行树状分组以及结构化数据导出 (JSON/CSV)。 - -## 激活时机 -- 当用户想要审核最近的系统登录和注销活动时。 -- 当以结构化格式检查系统重启历史时。 -- 当在 TTY 中对用户会话进行交互式分析时。 -- 当为安全报告或外部分析导出登录记录时。 - -## 核心原则与规范 -- **交互式浏览**: TTY 中的默认行为是记录的交互式选择器。 -- **树状视图**: 使用 `--tree` 按会话之前的重启事件进行分组。 -- **结构化格式**: 支持 `-j` (JSON) 和 `-c` (CSV) 以便于解析。 - -## 实战示例 - -### 交互式审核 -```bash -# 交互式浏览所有登录/注销记录 -x last -``` - -### 重启历史 -```bash -# 仅显示系统重启记录 -x last --reboot -``` - -### JSON 导出 -```bash -# 以 JSON 对象形式获取所有登录历史 -x last --json -``` - -### 分组视图 -```bash -# 以树状结构显示按系统重启分组的登录记录 -x last --tree -``` - -## 交付验证清单 -- [ ] 确认用户是否需要特定的格式(JSON, CSV, 树状)。 -- [ ] 验证是否需要按用户或重启进行过滤。 diff --git a/data/x-cmd/ll/SKILL.md b/data/x-cmd/ll/SKILL.md deleted file mode 100644 index 9c9d2a6..0000000 --- a/data/x-cmd/ll/SKILL.md +++ /dev/null @@ -1,37 +0,0 @@ ---- -name: ll -description: > - Search and list information about files, archives, and operating system resources using fzf. - Core Scenario: When the user needs an interactive way to explore file metadata or system resource summaries. -license: MIT ---- - -# ll - Interactive Resource Listing - -The `ll` module provides an interactive interface, primarily leveraging `fzf`, to search and list information about the file system and key operating system resources like CPU, memory, and processes. - -## When to Activate -- When the user wants to search for files or directories interactively using `fzf`. -- When performing a quick interactive check of system CPU, memory, or process states. - -## Core Principles & Rules -- **Interactive Search**: Use the `--fzfapp` (or default `x ll`) to search through files and system data. -- **Resource Shortcuts**: Supports prefixed subcommands (e.g., `:cpu`, `:mem`) for quick system snapshots. - -## Patterns & Examples - -### Interactive File Search -```bash -# Start an interactive fzf session to browse files and directories -x ll -``` - -### Quick System Check -```bash -# View current memory or process state via ll -x ll :mem -x ll :ps -``` - -## Checklist -- [ ] Confirm if the user needs an interactive search or a specific resource snapshot. diff --git a/data/x-cmd/ll/SKILL.zh.md b/data/x-cmd/ll/SKILL.zh.md deleted file mode 100644 index 685067d..0000000 --- a/data/x-cmd/ll/SKILL.zh.md +++ /dev/null @@ -1,37 +0,0 @@ ---- -name: ll -description: > - 使用 fzf 搜索并列出有关文件、压缩包和操作系统资源的信息。 - 核心场景:当用户需要以交互方式探索文件元数据或系统资源摘要时。 -license: MIT ---- - -# ll - 交互式资源列表 - -`ll` 模块提供了一个交互式界面(主要利用 `fzf`),用于搜索和列出有关文件系统以及关键操作系统资源(如 CPU、内存和进程)的信息。 - -## 激活时机 -- 当用户想要使用 `fzf` 交互式地搜索文件或目录时。 -- 当对系统 CPU、内存或进程状态执行快速交互式检查时。 - -## 核心原则与规范 -- **交互式搜索**: 使用 `--fzfapp`(或默认的 `x ll`)搜索文件和系统数据。 -- **资源快捷方式**: 支持前缀子命令(例如 `:cpu`, `:mem`)以获取快速的系统快照。 - -## 实战示例 - -### 交互式文件搜索 -```bash -# 启动交互式 fzf 会话以浏览文件和目录 -x ll -``` - -### 快速系统检查 -```bash -# 通过 ll 查看当前内存或进程状态 -x ll :mem -x ll :ps -``` - -## 交付验证清单 -- [ ] 确认用户需要的是交互式搜索还是特定的资源快照。 diff --git a/data/x-cmd/llmf/SKILL.md b/data/x-cmd/llmf/SKILL.md deleted file mode 100644 index cef7a21..0000000 --- a/data/x-cmd/llmf/SKILL.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: llmf -description: > - Run local LLMs with zero dependencies using llamafile, supporting API servers, CLI chat, and model management. - Core Scenario: When the user wants to run high-performance local models (GGUF/llamafile) without external dependencies or cloud APIs. -license: MIT ---- - -# llmf - Zero-Dependency Local LLM Runner - -The `llmf` module leverages llamafile technology to run large language models locally with zero setup. It provides a full suite of tools for serving, chatting, and managing models. - -## When to Activate -- When the user wants to run an AI model locally without cloud API access. -- When setting up a local OpenAI-compatible API server (`serve`). -- When performing fast, one-off text generation tasks via CLI. -- When managing local GGUF or llamafile models (download, import, tokenize). - -## Core Principles & Rules -- **Zero-Dependency**: Emphasize that models run locally without external runtimes. -- **Compatibility**: The `serve` command provides an OpenAI-compatible HTTP interface. -- **Resource Management**: Models are stored in `~/.x-cmd/data/llmf/model/`. - -## Additional Scenarios -- **Token Analysis**: Use `tokenize` to breakdown text into token details. -- **Headless Server**: Start the API server without opening a browser using `--nobrowser`. - -## Patterns & Examples - -### Start local API Server -```bash -# Run a specific model as an OpenAI-compatible server -x llmf serve -m llava/v1.5-7b/q4_k.gguf --nobrowser -``` - -### Fast Text Generation -```bash -# Execute a single prompt and output the result -x llmf cli -p 'Write a concise summary of the Python GIL' -``` - -### Download Model -```bash -# Pull a specific model from the library -x llmf model download llava/v1.5-7b/q4_k.gguf -``` - -## Checklist -- [ ] Ensure the user has enough disk space for local models. -- [ ] Verify if the chosen model format (GGUF/llamafile) is supported. -- [ ] Check if the local API port is available if running `serve`. diff --git a/data/x-cmd/llmf/SKILL.zh.md b/data/x-cmd/llmf/SKILL.zh.md deleted file mode 100644 index 7d64074..0000000 --- a/data/x-cmd/llmf/SKILL.zh.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: llmf -description: > - 使用 llamafile 零依赖运行本地大模型,支持 API 服务器、CLI 对话和模型管理。 - 核心场景:当用户想要在没有外部依赖或云端 API 的情况下运行高性能本地模型 (GGUF/llamafile) 时。 -license: MIT ---- - -# llmf - 零依赖本地大模型运行工具 - -`llmf` 模块利用 llamafile 技术,无需设置即可在本地运行大语言模型。它提供了一套完整的工具,用于服务托管、对话和模型管理。 - -## 激活时机 -- 当用户想要在没有云端 API 访问权限的情况下本地运行 AI 模型时。 -- 当设置本地 OpenAI 兼容的 API 服务器 (`serve`) 时。 -- 当通过 CLI 执行快速的一次性文本生成任务时。 -- 当管理本地 GGUF 或 llamafile 模型(下载、导入、分词)时。 - -## 核心原则与规范 -- **零依赖**: 强调模型在没有外部运行时的情况下本地运行。 -- **兼容性**: `serve` 命令提供 OpenAI 兼容的 HTTP 接口。 -- **资源管理**: 模型存储在 `~/.x-cmd/data/llmf/model/` 目录中。 - -## 补充场景 -- **Token 分析**: 使用 `tokenize` 将文本分解为 Token 详情。 -- **无头服务器**: 使用 `--nobrowser` 启动 API 服务而不打开浏览器。 - -## 实战示例 - -### 启动本地 API 服务 -```bash -# 将特定模型作为 OpenAI 兼容服务器运行 -x llmf serve -m llava/v1.5-7b/q4_k.gguf --nobrowser -``` - -### 快速文本生成 -```bash -# 执行单次提示并输出结果 -x llmf cli -p '简要总结 Python 的 GIL' -``` - -### 下载模型 -```bash -# 从库中拉取特定模型 -x llmf model download llava/v1.5-7b/q4_k.gguf -``` - -## 交付验证清单 -- [ ] 确保用户有足够的磁盘空间存放本地模型。 -- [ ] 验证所选模型格式 (GGUF/llamafile) 是否受支持。 -- [ ] 如果运行 `serve`,检查本地 API 端口是否可用。 diff --git a/data/x-cmd/lms/SKILL.md b/data/x-cmd/lms/SKILL.md deleted file mode 100644 index 60a0026..0000000 --- a/data/x-cmd/lms/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: lms -description: > - CLI module for LM Studio, enabling terminal-based chat and local LLM management. - Core Scenario: When the user wants to interact with locally hosted models in LM Studio via the command line. -license: MIT ---- - -# lms - LM Studio CLI Enhancement - -The `lms` module provides a CLI interface for LM Studio, allowing users to chat with local models and manage configurations directly from the terminal. - -## When to Activate -- When the user wants to chat with models running in LM Studio. -- When managing local LM Studio configurations and session defaults. -- When performing terminal-based interaction with local AI services. - -## Core Principles & Rules -- **Integration**: Designed to work alongside the LM Studio desktop application. -- **Subcommand Transparency**: Use `--runcmd` to access original `lms` command features if needed. - -## Patterns & Examples - -### Chat with Local Model -```bash -# Start a chat session with the model active in LM Studio -x lms chat -``` - -### Initialize Config -```bash -# Set up default parameters for LM Studio interaction -x lms init -``` - -## Checklist -- [ ] Ensure LM Studio is running and the local server is active. -- [ ] Verify if specific session defaults need to be set via `x lms --cur`. diff --git a/data/x-cmd/lms/SKILL.zh.md b/data/x-cmd/lms/SKILL.zh.md deleted file mode 100644 index 7ba5145..0000000 --- a/data/x-cmd/lms/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: lms -description: > - LM Studio 命令行模块,支持基于终端的聊天和本地大模型管理。 - 核心场景:当用户想要通过命令行与 LM Studio 中托管的本地模型进行交互时。 -license: MIT ---- - -# lms - LM Studio 命令行增强 - -`lms` 模块为 LM Studio 提供命令行界面,允许用户直接从终端与本地模型聊天并管理配置。 - -## 激活时机 -- 当用户想要与 LM Studio 中运行的模型聊天时。 -- 当管理本地 LM Studio 配置和会话默认值时。 -- 当通过终端与本地 AI 服务进行交互时。 - -## 核心原则与规范 -- **集成性**: 旨在与 LM Studio 桌面应用程序配合工作。 -- **子命令透明度**: 如有需要,使用 `--runcmd` 访问原始 `lms` 命令的功能。 - -## 实战示例 - -### 与本地模型聊天 -```bash -# 与 LM Studio 中活跃的模型开始聊天会话 -x lms chat -``` - -### 初始化配置 -```bash -# 设置与 LM Studio 交互的默认参数 -x lms init -``` - -## 交付验证清单 -- [ ] 确保 LM Studio 正在运行且本地服务器已激活。 -- [ ] 验证是否需要通过 `x lms --cur` 设置特定的会话默认值。 diff --git a/data/x-cmd/ls/SKILL.md b/data/x-cmd/ls/SKILL.md deleted file mode 100644 index 0460b57..0000000 --- a/data/x-cmd/ls/SKILL.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: ls -description: > - Enhanced ls command providing unified access to system resources like CPU, memory, and processes. - Core Scenario: When the user needs a quick summary of files and system resource states via enhanced subcommands. -license: MIT ---- - -# ls - Enhanced Resource Listing - -The `ls` module extends the traditional file listing command to provide rapid access to various system resources (CPU, Memory, Processes, PATH) through a unified interface. - -## When to Activate -- When the user wants to view file info alongside system resource summaries. -- When performing a quick check of PATH variables or process lists within a listing workflow. -- When using an interactive app (`:app`) to browse file metadata. - -## Core Principles & Rules -- **Resource Focused**: Use prefixed subcommands (starting with `:`) to target specific system data. -- **Interactive UI**: Use `:app` for a richer metadata exploration than standard listing. - -## Patterns & Examples - -### System Summary -```bash -# Quickly view CPU and Memory info using enhanced ls -x ls :cpu -x ls :mem -``` - -### Environment Check -```bash -# List all directories currently in the PATH variable -x ls :path -``` - -### Metadata Explorer -```bash -# Open an interactive viewer for files and directory information -x ls :app -``` - -## Checklist -- [ ] Confirm if the user needs file listing or a system resource summary. -- [ ] Verify if an interactive view (`:app`) is preferred over static output. diff --git a/data/x-cmd/ls/SKILL.zh.md b/data/x-cmd/ls/SKILL.zh.md deleted file mode 100644 index 738f616..0000000 --- a/data/x-cmd/ls/SKILL.zh.md +++ /dev/null @@ -1,45 +0,0 @@ ---- -name: ls -description: > - 增强型 ls 命令,提供对 CPU、内存和进程等系统资源的统一访问。 - 核心场景:当用户需要通过增强型子命令快速获取文件和系统资源状态摘要时。 -license: MIT ---- - -# ls - 增强型资源列表 - -`ls` 模块扩展了传统的文件列表命令,通过统一的界面提供对各种系统资源(CPU、内存、进程、PATH)的快速访问。 - -## 激活时机 -- 当用户想要在查看文件信息的同时查看系统资源摘要时。 -- 当在列表工作流中快速检查 PATH 变量或进程列表时。 -- 当使用交互式应用 (`:app`) 浏览文件元数据时。 - -## 核心原则与规范 -- **资源关注**: 使用以 `:` 开头的前缀子命令来针对特定的系统数据。 -- **交互式 UI**: 使用 `:app` 获得比标准列表更丰富的元数据探索。 - -## 实战示例 - -### 系统摘要 -```bash -# 使用增强型 ls 快速查看 CPU 和内存信息 -x ls :cpu -x ls :mem -``` - -### 环境检查 -```bash -# 列出 PATH 变量中当前引入的所有目录 -x ls :path -``` - -### 元数据浏览器 -```bash -# 打开文件和目录信息的交互式查看器 -x ls :app -``` - -## 交付验证清单 -- [ ] 确认用户需要的是文件列表还是系统资源摘要。 -- [ ] 验证交互式视图 (`:app`) 是否优于静态输出。 diff --git a/data/x-cmd/lsio/SKILL.md b/data/x-cmd/lsio/SKILL.md deleted file mode 100644 index 15c8565..0000000 --- a/data/x-cmd/lsio/SKILL.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: lsio -description: > - Manage and deploy common containers from the LinuxServer.io ecosystem. - Core Scenario: When the user needs to quickly set up popular open-source services like code-server or filebrowser. -license: MIT ---- - -# lsio - LinuxServer.io Container Management - -The `lsio` module provides a simplified way to manage and run popular containerized applications provided by LinuxServer.io, ensuring easy setup and consistent configuration. - -## When to Activate -- When setting up a code-server (VS Code in browser) or filebrowser instance. -- When managing home automation or media server containers from LSIO. - -## Patterns & Examples - -### Run Code Server -```bash -# Start a code-server container for browser-based development -x lsio code-server run -``` - -## Checklist -- [ ] Confirm Docker is available. -- [ ] Verify the target application name from the LSIO catalog. diff --git a/data/x-cmd/lsio/SKILL.zh.md b/data/x-cmd/lsio/SKILL.zh.md deleted file mode 100644 index 39b1e3c..0000000 --- a/data/x-cmd/lsio/SKILL.zh.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: lsio -description: > - 管理和部署来自 LinuxServer.io 生态系统的常用容器。 - 核心场景:当用户需要快速设置 code-server 或 filebrowser 等流行的开源服务时。 -license: MIT ---- - -# lsio - LinuxServer.io 容器管理 - -`lsio` 模块提供了一种简化的方法来管理和运行由 LinuxServer.io 提供的流行容器化应用程序,确保简单的设置和一致的配置。 - -## 激活时机 -- 当设置 code-server(浏览器中的 VS Code)或 filebrowser 实例时。 -- 当管理来自 LSIO 的家庭自动化或媒体服务器容器时。 - -## 实战示例 - -### 运行 Code Server -```bash -# 启动一个 code-server 容器以进行基于浏览器的开发 -x lsio code-server run -``` - -## 交付验证清单 -- [ ] 确认 Docker 可用。 -- [ ] 验证 LSIO 目录中的目标应用程序名称。 diff --git a/data/x-cmd/lsof/SKILL.md b/data/x-cmd/lsof/SKILL.md deleted file mode 100644 index 42623e5..0000000 --- a/data/x-cmd/lsof/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: lsof -description: > - Enhanced lsof interface for viewing open files with TUI support and structured data exports (TSV, CSV). - Core Scenario: When the user needs to identify which files are open by which processes, supporting interactive search. -license: MIT ---- - -# lsof - List Open Files Enhancement - -The `lsof` module provides a more intuitive way to interact with the system's open file list. It supports an interactive TUI for exploration and allows exporting results in common data formats. - -## When to Activate -- When investigating file locking issues or identifying process ownership of files. -- When performing interactive searches of open files using FZF. -- When exporting open file data to CSV or TSV for script processing. -- When needing thread-level details (`--task`) for open files. - -## Core Principles & Rules -- **Interactive TUI**: Automatically displays a TUI when connected to a terminal; otherwise, defaults to TSV. -- **Data-Centric**: Use `--csv` or `--tsv` when the output is intended for automation or external analysis. -- **Searchable**: Leverage the `fz` subcommand for fast interactive filtering. - -## Patterns & Examples - -### Interactive Search -```bash -# Interactively search and view open files via FZF -x lsof fz -``` - -### Export to CSV -```bash -# Output current open files as a CSV for processing -x lsof --csv -``` - -### View Thread Info -```bash -# Display task (thread) IDs and names for open files -x lsof --task -``` - -## Checklist -- [ ] Confirm if the user needs an interactive view or a static data export. -- [ ] Verify if thread-level information is required. -- [ ] Ensure the correct export format (CSV/TSV) is selected if needed. diff --git a/data/x-cmd/lsof/SKILL.zh.md b/data/x-cmd/lsof/SKILL.zh.md deleted file mode 100644 index 67e3483..0000000 --- a/data/x-cmd/lsof/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: lsof -description: > - 增强型 lsof 接口,用于查看打开的文件,支持 TUI 和结构化数据导出 (TSV, CSV)。 - 核心场景:当用户需要识别哪些进程打开了哪些文件时,支持交互式搜索。 -license: MIT ---- - -# lsof - 列出打开文件增强工具 - -`lsof` 模块提供了一种更直观的方式来与系统的打开文件列表进行交互。它支持用于探索的交互式 TUI,并允许以常见数据格式导出结果。 - -## 激活时机 -- 当调查文件锁定问题或识别文件的进程所有权时。 -- 当使用 FZF 对打开的文件执行交互式搜索时。 -- 当为脚本处理导出打开的文件数据到 CSV 或 TSV 时。 -- 当需要打开文件的线程级详情 (`--task`) 时。 - -## 核心原则与规范 -- **交互式 TUI**: 连接到终端时自动显示 TUI;否则,默认为 TSV。 -- **以数据为中心**: 当输出打算用于自动化或外部分析时,使用 `--csv` 或 `--tsv`。 -- **可搜索性**: 利用 `fz` 子命令进行快速的交互式过滤。 - -## 实战示例 - -### 交互式搜索 -```bash -# 通过 FZF 交互式搜索并查看打开的文件 -x lsof fz -``` - -### 导出为 CSV -```bash -# 将当前打开的文件输出为 CSV 以供处理 -x lsof --csv -``` - -### 查看线程信息 -```bash -# 显示打开文件的任务 (线程) ID 和名称 -x lsof --task -``` - -## 交付验证清单 -- [ ] 确认用户需要的是交互式视图还是静态数据导出。 -- [ ] 验证是否需要线程级信息。 -- [ ] 确保如果需要,选择了正确的导出格式 (CSV/TSV)。 diff --git a/data/x-cmd/lua/SKILL.md b/data/x-cmd/lua/SKILL.md deleted file mode 100644 index 479c8ff..0000000 --- a/data/x-cmd/lua/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: lua -description: > - Enhanced Lua development module, supporting project initialization, module installation via Luarocks, and static compilation. - Core Scenario: When the user needs to set up Lua projects, manage libraries, or compile Lua scripts to static binaries. -license: MIT ---- - -# lua - Enhanced Lua Development - -The `lua` module provides a comprehensive suite of tools for Lua developers, simplifying project setup, dependency management, and script execution. - -## When to Activate -- When initializing a new Lua project structure (`init`). -- When installing Lua modules and libraries using Luarocks integration (`install`). -- When formatting, checking, or linting Lua source code. -- When compiling Lua scripts into standalone static binaries. - -## Core Principles & Rules -- **Luarocks Integration**: Use the `install` (or `i`) subcommand to manage dependencies easily. -- **Static Compilation**: Leverage the `static` subcommand for creating zero-dependency Lua executables. -- **Code Quality**: Supports `check` and `format` for maintaining Lua code standards. - -## Patterns & Examples - -### Install Library -```bash -# Install the 'lua-cjson' module using the integrated Luarocks -x lua i lua-cjson -``` - -### Static Build -```bash -# Compile a Lua script into a single static binary -x lua static main.lua -``` - -### Initialize Project -```bash -# Set up a new Lua project environment -x lua init -``` - -## Checklist -- [ ] Confirm if the user needs to manage dependencies or execute a script. -- [ ] Verify the target platform for static compilation. diff --git a/data/x-cmd/lua/SKILL.zh.md b/data/x-cmd/lua/SKILL.zh.md deleted file mode 100644 index 902bdac..0000000 --- a/data/x-cmd/lua/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: lua -description: > - 增强型 Lua 开发模块,支持项目初始化、通过 Luarocks 安装模块以及静态编译。 - 核心场景:当用户需要设置 Lua 项目、管理库或将 Lua 脚本编译为静态二进制文件时。 -license: MIT ---- - -# lua - 增强型 Lua 开发工具 - -`lua` 模块为 Lua 开发者提供了一套全面的工具,简化了项目设置、依赖项管理和脚本执行。 - -## 激活时机 -- 当初始化一个新的 Lua 项目结构时 (`init`)。 -- 当使用集成的 Luarocks 安装 Lua 模块和库时 (`install`)。 -- 当格式化、检查或对 Lua 源代码进行 lint 时。 -- 当将 Lua 脚本编译为独立的静态二进制文件时。 - -## 核心原则与规范 -- **Luarocks 集成**: 使用 `install` (或 `i`) 子命令轻松管理依赖项。 -- **静态编译**: 利用 `static` 子命令创建零依赖的 Lua 可执行文件。 -- **代码质量**: 支持使用 `check` 和 `format` 维护 Lua 代码标准。 - -## 实战示例 - -### 安装库 -```bash -# 使用集成的 Luarocks 安装 'lua-cjson' 模块 -x lua i lua-cjson -``` - -### 静态构建 -```bash -# 将 Lua 脚本编译为单个静态二进制文件 -x lua static main.lua -``` - -### 初始化项目 -```bash -# 设置一个新的 Lua 项目环境 -x lua init -``` - -## 交付验证清单 -- [ ] 确认用户需要管理依赖项还是执行脚本。 -- [ ] 验证静态编译的目标平台。 diff --git a/data/x-cmd/luarocks/SKILL.md b/data/x-cmd/luarocks/SKILL.md deleted file mode 100644 index 06fff6c..0000000 --- a/data/x-cmd/luarocks/SKILL.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: luarocks -description: > - Enhanced interface for Luarocks, the package manager for Lua modules. - Core Scenario: When the user needs to install, manage, or search for Lua dependencies. -license: MIT ---- - -# luarocks - Lua Package Management - -The `luarocks` module provides a streamlined CLI for managing Lua modules, integrating with x-cmd's package system for easy installation. - -## When to Activate -- When installing or removing Lua packages (Rocks). -- When searching the Luarocks repository for modules. - -## Patterns & Examples - -### Install Module -```bash -# Install a specific Lua module -x luarocks install luasocket -``` - -## Checklist -- [ ] Confirm the module name. -- [ ] Verify if the Lua runtime is correctly set up. diff --git a/data/x-cmd/luarocks/SKILL.zh.md b/data/x-cmd/luarocks/SKILL.zh.md deleted file mode 100644 index 07ddde7..0000000 --- a/data/x-cmd/luarocks/SKILL.zh.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: luarocks -description: > - Luarocks 的增强接口,Lua 模块的包管理器。 - 核心场景:当用户需要安装、管理或搜索 Lua 依赖项时。 -license: MIT ---- - -# luarocks - Lua 包管理 - -`luarocks` 模块提供了一个简化的命令行界面来管理 Lua 模块,并与 x-cmd 的包系统集成以便于安装。 - -## 激活时机 -- 当安装或移除 Lua 软件包 (Rocks) 时。 -- 当在 Luarocks 仓库中搜索模块时。 - -## 实战示例 - -### 安装模块 -```bash -# 安装特定的 Lua 模块 -x luarocks install luasocket -``` - -## 交付验证清单 -- [ ] 确认模块名称。 -- [ ] 验证 Lua 运行时是否已正确设置。 diff --git a/data/x-cmd/mac/SKILL.md b/data/x-cmd/mac/SKILL.md deleted file mode 100644 index 7375b82..0000000 --- a/data/x-cmd/mac/SKILL.md +++ /dev/null @@ -1,49 +0,0 @@ ---- -name: mac -description: > - Integrated utility for macOS users to manage system features via CLI and scripts. - Core Scenario: When the user needs to configure TouchID for sudo, use application sandboxes, or control macOS system settings. -license: MIT ---- - -# mac - macOS CLI Utilities - -The `mac` module provides a comprehensive set of tools for macOS users, simplifying system administration, automation, and remote management through a unified CLI. - -## When to Activate -- When the user wants to enable TouchID for sudo authentication. -- When managing macOS-specific features like Dock, Launchpad, or Wallpapers. -- When controlling system volume, battery info, or performing power operations (shutdown, lock, sleep). -- When using the `sandbox` feature to restrict application access to specific directories. -- When managing Apple Notes or Reminders from the terminal. - -## Core Principles & Rules -- **Automation-Friendly**: Designed for use in shell scripts to automate macOS configuration. -- **Safety Levels**: Use appropriate levels in the `sandbox` subcommand (`-0` to `-9`) to balance permission and security. -- **Alias Support**: Use `alias enable` to set up `m` as a shortcut for `x mac`. - -## Patterns & Examples - -### Enable TouchID for Sudo -```bash -# Allow using TouchID for sudo verification -x mac tidsudo enable -``` - -### Application Sandboxing -```bash -# Run Claude with restricted access to sensitive user folders -x mac sb -9 -d "$HOME/.ssh" -d "$HOME/Library" claude -``` - -### System Control -```bash -# Set system volume to 50% and lock the screen -x mac vol set 50 -x mac lock -``` - -## Checklist -- [ ] Confirm if the command requires administrative (sudo) privileges. -- [ ] Verify the specific paths to be allowed or denied when using `sandbox`. -- [ ] Ensure the correct Apple ID or App ID is used for Store/App management. diff --git a/data/x-cmd/mac/SKILL.zh.md b/data/x-cmd/mac/SKILL.zh.md deleted file mode 100644 index 9f717d6..0000000 --- a/data/x-cmd/mac/SKILL.zh.md +++ /dev/null @@ -1,49 +0,0 @@ ---- -name: mac -description: > - 集成的 macOS 实用程序,用于通过 CLI 和脚本管理系统功能。 - 核心场景:当用户需要为 sudo 配置 TouchID、使用应用程序沙盒或控制 macOS 系统设置时。 -license: MIT ---- - -# mac - macOS 命令行实用工具 - -`mac` 模块为 macOS 用户提供了一套全面的工具,通过统一的 CLI 简化系统管理、自动化和远程管理。 - -## 激活时机 -- 当用户想要为 sudo 身份验证启用 TouchID 时。 -- 当管理 Dock、Launchpad 或壁纸等 macOS 特定功能时。 -- 当控制系统音量、电池信息或执行电源操作(关机、锁定、睡眠)时。 -- 当使用 `sandbox` 功能限制应用程序对特定目录的访问时。 -- 当从终端管理 Apple 备忘录或提醒事项时。 - -## 核心原则与规范 -- **自动化友好**: 旨在用于 Shell 脚本以自动执行 macOS 配置。 -- **安全级别**: 在 `sandbox` 子命令中使用合适的级别(`-0` 到 `-9`),以平衡权限和安全性。 -- **别名支持**: 使用 `alias enable` 将 `m` 设置为 `x mac` 的快捷方式。 - -## 实战示例 - -### 为 Sudo 启用 TouchID -```bash -# 允许使用 TouchID 进行 sudo 验证 -x mac tidsudo enable -``` - -### 应用程序沙盒化 -```bash -# 在限制访问敏感用户文件夹的情况下运行 Claude -x mac sb -9 -d "$HOME/.ssh" -d "$HOME/Library" claude -``` - -### 系统控制 -```bash -# 将系统音量设置为 50% 并锁定屏幕 -x mac vol set 50 -x mac lock -``` - -## 交付验证清单 -- [ ] 确认命令是否需要管理 (sudo) 权限。 -- [ ] 使用 `sandbox` 时,验证要允许或拒绝的具体路径。 -- [ ] 确保在进行商店/应用管理时使用了正确的 Apple ID 或 App ID。 diff --git a/data/x-cmd/man/SKILL.md b/data/x-cmd/man/SKILL.md deleted file mode 100644 index 3ee419f..0000000 --- a/data/x-cmd/man/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: man -description: > - Enhanced man page viewer with FZF integration and AI-powered command explanation. - Core Scenario: When the user needs to browse manual pages or requires an AI explanation of complex command options. -license: MIT ---- - -# man - Enhanced Manual Page Viewer - -The `man` module extends the standard system manual capabilities by adding interactive browsing via FZF and integrating with AI services like ManKier to explain complex command strings. - -## When to Activate -- When the user wants to search for man pages using keywords. -- When an interactive list of all system man pages is required. -- When the user needs an AI to explain the meaning of specific command flags (e.g., `tar -czvf`). -- When a quick community reference (TLDR style) is needed for a command via `x man :keyword`. - -## Core Principles & Rules -- **Interactive Search**: Use `--fzf` for a searchable TUI experience. -- **AI Explanation**: Leverage the `--explain` flag to breakdown complex commands into readable summaries. -- **Hybrid Support**: Support for accessing TLDR content directly from within the `man` command using the `:` prefix. - -## Patterns & Examples - -### Fuzzy Search Manuals -```bash -# Interactively choose from all system manual pages -x man --fzf -``` - -### AI Command Explanation -```bash -# Use AI to explain exactly what these tar flags do -x man --explain "tar -czvf" -``` - -### Quick TLDR Style -```bash -# View the simplified TLDR reference for ssh -x man :ssh -``` - -## Checklist -- [ ] Confirm if the user needs the full manual or just a quick explanation. -- [ ] Verify if the command string for explanation is complete. diff --git a/data/x-cmd/man/SKILL.zh.md b/data/x-cmd/man/SKILL.zh.md deleted file mode 100644 index 52252e6..0000000 --- a/data/x-cmd/man/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: man -description: > - 增强型手册页查看器,集成 FZF 和 AI 驱动的命令解释。 - 核心场景:当用户需要浏览手册页或需要 AI 对复杂命令选项进行解释时。 -license: MIT ---- - -# man - 增强型手册页查看器 - -`man` 模块通过添加 FZF 交互式浏览,并集成 ManKier 等 AI 服务来解释复杂的命令行字符串,扩展了标准的系统手册功能。 - -## 激活时机 -- 当用户想要使用关键词搜索手册页时。 -- 当需要系统所有手册页的交互式列表时。 -- 当用户需要 AI 解释特定命令标志(如 `tar -czvf`)的含义时。 -- 当需要通过 `x man :关键词` 快速查看命令的社区参考(TLDR 风格)时。 - -## 核心原则与规范 -- **交互式搜索**: 使用 `--fzf` 获得可搜索的 TUI 体验。 -- **AI 解释**: 利用 `--explain` 标志将复杂的命令分解为可读的摘要。 -- **混合支持**: 支持使用 `:` 前缀直接在 `man` 命令中访问 TLDR 内容。 - -## 实战示例 - -### 模糊搜索手册 -```bash -# 从所有系统手册页中交互式选择 -x man --fzf -``` - -### AI 命令解释 -```bash -# 使用 AI 准确解释这些 tar 标志的作用 -x man --explain "tar -czvf" -``` - -### 快速 TLDR 风格 -```bash -# 查看 ssh 的简化 TLDR 参考 -x man :ssh -``` - -## 交付验证清单 -- [ ] 确认用户需要完整的手册还是仅仅是快速解释。 -- [ ] 验证用于解释的命令字符串是否完整。 diff --git a/data/x-cmd/mankier/SKILL.md b/data/x-cmd/mankier/SKILL.md deleted file mode 100644 index 8880c51..0000000 --- a/data/x-cmd/mankier/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: mankier -description: > - Command-line client for ManKier.com, offering manual page queries and detailed command explanation. - Core Scenario: When the user needs to explain shell commands or fetch specific sections of online manual pages. -license: MIT ---- - -# mankier - Online Manual Page Queries & Explanation - -The `mankier` module provides a CLI interface for the ManKier.com API, allowing users to search, browse, and explain manual pages directly from the web without local installation. - -## When to Activate -- When the user needs a detailed breakdown of a shell command's flags and arguments. -- When searching for manual pages that are not installed on the local system. -- When retrieving specific sections (e.g., NAME, DESCRIPTION) of a manual page. -- When following cross-references between different manual pages. - -## Core Principles & Rules -- **Remote Accuracy**: Uses the latest online documentation from ManKier.com. -- **Granular Retrieval**: Supports fetching specific sections using the `section` subcommand. -- **Command breakdown**: Prioritize the `explain` subcommand for translating technical flags into human-readable text. - -## Patterns & Examples - -### Command Breakdown -```bash -# Explain the meaning of flags in a jq command -x mankier explain jq -cr -``` - -### Fetch Section -```bash -# Retrieve only the NAME section of the tar manual -x mankier section NAME tar -``` - -### Web Search Integration -```bash -# Search ManKier via DuckDuckGo for NVMe related pages -x mankier : nvme -``` - -## Checklist -- [ ] Confirm the command or flag the user needs explained. -- [ ] Verify if a specific section of the manual is required. -- [ ] Ensure the user is aware the data is fetched from an online source. diff --git a/data/x-cmd/mankier/SKILL.zh.md b/data/x-cmd/mankier/SKILL.zh.md deleted file mode 100644 index 77fe2f8..0000000 --- a/data/x-cmd/mankier/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: mankier -description: > - ManKier.com 的命令行客户端,提供手册页查询和详细的命令解释。 - 核心场景:当用户需要解释 Shell 命令或获取在线手册页的特定部分时。 -license: MIT ---- - -# mankier - 在线手册页查询与解释 - -`mankier` 模块为 ManKier.com API 提供了命令行界面,允许用户直接从 Web 搜索、浏览和解释手册页,无需本地安装。 - -## 激活时机 -- 当用户需要详细分解 Shell 命令的标志和参数时。 -- 当搜索本地系统中未安装的手册页时。 -- 当检索手册页的特定部分(如 NAME, DESCRIPTION)时。 -- 当跟踪不同手册页之间的交叉引用时。 - -## 核心原则与规范 -- **远程准确性**: 使用来自 ManKier.com 的最新在线文档。 -- **粒度检索**: 支持使用 `section` 子命令获取特定部分。 -- **命令分解**: 优先使用 `explain` 子命令将技术标志翻译为人机可读的文本。 - -## 实战示例 - -### 命令分解 -```bash -# 解释 jq 命令中标志的含义 -x mankier explain jq -cr -``` - -### 获取特定部分 -```bash -# 仅检索 tar 手册的 NAME 部分 -x mankier section NAME tar -``` - -### 网页搜索集成 -```bash -# 通过 DuckDuckGo 在 ManKier 网站搜索 NVMe 相关页面 -x mankier : nvme -``` - -## 交付验证清单 -- [ ] 确认用户需要解释的命令或标志。 -- [ ] 验证是否需要手册的特定部分。 -- [ ] 确保用户知晓数据是从在线源获取的。 diff --git a/data/x-cmd/mistral/SKILL.md b/data/x-cmd/mistral/SKILL.md deleted file mode 100644 index 6c279a8..0000000 --- a/data/x-cmd/mistral/SKILL.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: mistral -description: > - Integrate Mistral AI for efficient language modeling and text generation tasks. - Core Scenario: When the user wants to use Mistral models for chat or translation via CLI. -license: MIT ---- - -# mistral - Mistral AI Integration - -The `mistral` module provides a CLI interface for Mistral AI's services, supporting high-efficiency text generation and multi-language translation. - -## When to Activate -- When the user wants to chat with Mistral models. -- When the user needs to translate file content using Mistral's API. -- When checking or managing Mistral API keys. - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` for setup. -- **Alias Access**: Use the `@mistral` alias for fast interaction. -- **Model Exploration**: Use `model ls` to see all available Mistral models. - -## Patterns & Examples - -### Translate Files -```bash -# Use Mistral to translate a file to Chinese -@mistral --file ./document.en.md "Translate to Chinese" -``` - -### List Models -```bash -# View all supported Mistral models -x mistral model ls -``` - -## Checklist -- [ ] Ensure the Mistral API key is configured. -- [ ] Verify file paths when using the `--file` flag. diff --git a/data/x-cmd/mistral/SKILL.zh.md b/data/x-cmd/mistral/SKILL.zh.md deleted file mode 100644 index 01260bf..0000000 --- a/data/x-cmd/mistral/SKILL.zh.md +++ /dev/null @@ -1,39 +0,0 @@ ---- -name: mistral -description: > - 集成 Mistral AI,用于高效的语言建模和文本生成任务。 - 核心场景:当用户想要通过 CLI 使用 Mistral 模型进行聊天或翻译时。 -license: MIT ---- - -# mistral - Mistral AI 集成 - -`mistral` 模块为 Mistral AI 的服务提供命令行界面,支持高效的文本生成和多语言翻译。 - -## 激活时机 -- 当用户想要与 Mistral 模型聊天时。 -- 当用户需要使用 Mistral API 翻译文件内容时。 -- 当检查或管理 Mistral API 密钥时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 进行设置。 -- **别名访问**: 使用 `@mistral` 别名进行快速交互。 -- **探索模型**: 使用 `model ls` 查看所有可用的 Mistral 模型。 - -## 实战示例 - -### 翻译文件 -```bash -# 使用 Mistral 将文件翻译为中文 -@mistral --file ./document.en.md "翻译为中文" -``` - -### 列出模型 -```bash -# 查看所有支持的 Mistral 模型 -x mistral model ls -``` - -## 交付验证清单 -- [ ] 确保已配置 Mistral API 密钥。 -- [ ] 使用 `--file` 标志时验证文件路径。 diff --git a/data/x-cmd/moonshot/SKILL.md b/data/x-cmd/moonshot/SKILL.md deleted file mode 100644 index 77076b1..0000000 --- a/data/x-cmd/moonshot/SKILL.md +++ /dev/null @@ -1,43 +0,0 @@ ---- -name: moonshot -description: > - Integrate Moonshot AI (Kimi) for chat and file-based context processing. - Core Scenario: When the user wants to use Moonshot's large context capabilities for chat or file translation. -license: MIT ---- - -# moonshot - Moonshot AI (Kimi) Integration - -The `moonshot` module provides access to Moonshot AI's services, known for their large context window and strong performance in Chinese language tasks. - -## When to Activate -- When the user wants to chat with the Kimi model. -- When the user needs to analyze or translate files using Moonshot's context. -- When checking Moonshot account balance or managing file uploads. - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` for setup. -- **Alias Support**: Use the `@kimi` alias for faster access. -- **Context Handling**: Leverage the large context window by attaching files with `--file`. - -## Additional Scenarios -- **File Management**: Use the `file` subcommand to manage documents uploaded to Moonshot. -- **Balance Check**: Monitor credits using `x moonshot balance`. - -## Patterns & Examples - -### Translate Files with Kimi -```bash -# Use the @kimi alias to translate local files -@kimi --file ./content.en.md "Translate to Chinese" -``` - -### List Available Models -```bash -# View models supported by Moonshot -x moonshot model ls -``` - -## Checklist -- [ ] Ensure the Moonshot API key is configured. -- [ ] Verify that files attached exist and are readable. diff --git a/data/x-cmd/moonshot/SKILL.zh.md b/data/x-cmd/moonshot/SKILL.zh.md deleted file mode 100644 index c01a2ec..0000000 --- a/data/x-cmd/moonshot/SKILL.zh.md +++ /dev/null @@ -1,43 +0,0 @@ ---- -name: moonshot -description: > - 集成 Moonshot AI (Kimi),用于聊天和基于文件的上下文处理。 - 核心场景:当用户想要利用 Moonshot 的大上下文能力进行聊天或文件翻译时。 -license: MIT ---- - -# moonshot - Moonshot AI (Kimi) 集成 - -`moonshot` 模块提供对 Moonshot AI 服务的访问,该公司以其大上下文窗口和在中文任务中的强劲表现而闻名。 - -## 激活时机 -- 当用户想要与 Kimi 模型聊天时。 -- 当用户需要利用 Moonshot 的上下文分析或翻译文件时。 -- 当检查 Moonshot 账户余额或管理文件上传时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 进行设置。 -- **别名支持**: 使用 `@kimi` 别名进行更快速的访问。 -- **上下文处理**: 通过使用 `--file` 附加文件来利用其大上下文窗口。 - -## 补充场景 -- **文件管理**: 使用 `file` 子命令管理上传到 Moonshot 的文档。 -- **余额检查**: 使用 `x moonshot balance` 监控信用额度。 - -## 实战示例 - -### 使用 Kimi 翻译文件 -```bash -# 使用 @kimi 别名翻译本地文件 -@kimi --file ./content.en.md "翻译为中文" -``` - -### 列出可用模型 -```bash -# 查看 Moonshot 支持的模型 -x moonshot model ls -``` - -## 交付验证清单 -- [ ] 确保已配置 Moonshot API 密钥。 -- [ ] 验证附加的文件是否存在且可读。 diff --git a/data/x-cmd/mosh/SKILL.md b/data/x-cmd/mosh/SKILL.md deleted file mode 100644 index de7c2a6..0000000 --- a/data/x-cmd/mosh/SKILL.md +++ /dev/null @@ -1,40 +0,0 @@ ---- -name: mosh -description: > - Enhanced interface for Mosh (Mobile Shell), a robust SSH replacement for unstable networks. - Core Scenario: When the user needs a persistent, roaming remote terminal connection over UDP. -license: MIT ---- - -# mosh - Robust Remote Shell (UDP) - -The `mosh` module provides an enhanced wrapper for the Mobile Shell, ensuring availability by automatically downloading it via pixi if it is missing from the system. It is ideal for roaming and high-latency connections. - -## When to Activate -- When connecting to remote servers over unstable or high-latency networks (e.g., cellular data). -- When the user needs a connection that persists through network changes (roaming). -- When seeking an interactive SSH alternative with local echo and line editing features. - -## Core Principles & Rules -- **UDP Requirement**: Uses UDP ports (default 60000-61000) instead of TCP. -- **SSH dependency**: Requires a functioning SSH command to establish the initial handshake. -- **Cross-Platform**: Automatically handles setup via `pixi` on supported platforms (macOS/Linux). - -## Patterns & Examples - -### Basic Connection -```bash -# Connect to a remote server using Mosh -x mosh user@remote-host.com -``` - -### Specific Port and SSH Command -```bash -# Use a custom SSH port and specific Mosh UDP port -x mosh -p 80 --ssh="ssh -p 2222" root@server.com -``` - -## Checklist -- [ ] Ensure the remote server also has `mosh-server` installed. -- [ ] Verify that UDP ports (60000-61000) are open on the remote firewall. -- [ ] Confirm the target user and hostname. diff --git a/data/x-cmd/mosh/SKILL.zh.md b/data/x-cmd/mosh/SKILL.zh.md deleted file mode 100644 index 7a36f60..0000000 --- a/data/x-cmd/mosh/SKILL.zh.md +++ /dev/null @@ -1,40 +0,0 @@ ---- -name: mosh -description: > - Mosh (Mobile Shell) 的增强型接口,这是一款适用于不稳定网络的稳健 SSH 替代品。 - 核心场景:当用户需要通过 UDP 进行持久的、可漫游的远程终端连接时。 -license: MIT ---- - -# mosh - 稳健的远程 Shell (UDP) - -`mosh` 模块为 Mobile Shell 提供了一个增强型包装器,通过在系统缺失时利用 pixi 自动下载,确保了工具的可获得性。它是漫游和高延迟连接的理想选择。 - -## 激活时机 -- 当通过不稳定或高延迟的网络(如蜂窝数据)连接到远程服务器时。 -- 当用户需要一个能在网络更替(漫游)过程中保持持久的连接时。 -- 当寻求具有本地回显和行编辑功能的交互式 SSH 替代品时。 - -## 核心原则与规范 -- **UDP 要求**: 使用 UDP 端口(默认 60000-61000)而不是 TCP。 -- **SSH 依赖**: 需要一个正常运行的 SSH 命令来建立初始握手。 -- **跨平台**: 在受支持的平台(macOS/Linux)上自动通过 `pixi` 处理设置。 - -## 实战示例 - -### 基础连接 -```bash -# 使用 Mosh 连接到远程服务器 -x mosh user@remote-host.com -``` - -### 特定端口与 SSH 命令 -```bash -# 使用自定义 SSH 端口和特定的 Mosh UDP 端口 -x mosh -p 80 --ssh="ssh -p 2222" root@server.com -``` - -## 交付验证清单 -- [ ] 确保远程服务器也安装了 `mosh-server`。 -- [ ] 验证远程防火墙已开放 UDP 端口 (60000-61000)。 -- [ ] 确认目标用户和主机名。 diff --git a/data/x-cmd/nets/SKILL.md b/data/x-cmd/nets/SKILL.md deleted file mode 100644 index 49b2101..0000000 --- a/data/x-cmd/nets/SKILL.md +++ /dev/null @@ -1,40 +0,0 @@ ---- -name: nets -description: > - Enhanced netstat interface for viewing internet connections, routing tables, and network statistics. - Core Scenario: When the user needs an interactive way to audit network sockets (TCP/UDP) or analyze network tables. -license: MIT ---- - -# nets - Enhanced Network Statistics - -The `nets` module provides a user-friendly way to display network statistics, improving upon traditional `netstat`. It supports interactive views and structured data formats like CSV and TSV. - -## When to Activate -- When the user wants to audit active TCP/UDP connections (internet table). -- When analyzing the system's routing table in a readable format. -- When needing structured network data (CSV/TSV) for monitoring scripts. -- When performing a quick interactive overview of all network tables. - -## Core Principles & Rules -- **Interactive Browsing**: Default mode allows users to select which network table to view. -- **Structured Formats**: Support for `--csv` and `--tsv` for integration with other data processing tools. -- **Refresh Control**: Use `update` to ensure the table data is current. - -## Patterns & Examples - -### View Connections -```bash -# Show all active internet (TCP/UDP) connections -x nets view internet -``` - -### Export Routing Table -```bash -# Output the routing table as a TSV for analysis -x nets view --tsv route -``` - -## Checklist -- [ ] Confirm the specific network table (internet, route, etc.) to be viewed. -- [ ] Verify if the output needs to be structured (CSV/TSV) for automation. diff --git a/data/x-cmd/nets/SKILL.zh.md b/data/x-cmd/nets/SKILL.zh.md deleted file mode 100644 index 2cd1933..0000000 --- a/data/x-cmd/nets/SKILL.zh.md +++ /dev/null @@ -1,40 +0,0 @@ ---- -name: nets -description: > - 增强型 netstat 接口,用于查看互联网连接、路由表和网络统计信息。 - 核心场景:当用户需要以交互方式审核网络套接字 (TCP/UDP) 或分析网络表时。 -license: MIT ---- - -# nets - 增强型网络统计 - -`nets` 模块提供了一种用户友好的方式来显示网络统计信息,是对传统 `netstat` 的改进。它支持交互式视图和结构化数据格式(如 CSV 和 TSV)。 - -## 激活时机 -- 当用户想要审核活跃的 TCP/UDP 连接(互联网表)时。 -- 当以可读格式分析系统的路由表时。 -- 当为监控脚本需要结构化网络数据 (CSV/TSV) 时。 -- 当对所有网络表执行快速交互式概览时。 - -## 核心原则与规范 -- **交互式浏览**: 默认模式允许用户选择要查看的网络表。 -- **结构化格式**: 支持 `--csv` 和 `--tsv` 以便于与其他数据处理工具集成。 -- **刷新控制**: 使用 `update` 确保表数据是最新的。 - -## 实战示例 - -### 查看连接 -```bash -# 显示所有活跃的互联网 (TCP/UDP) 连接 -x nets view internet -``` - -### 导出路由表 -```bash -# 将路由表输出为 TSV 以供分析 -x nets view --tsv route -``` - -## 交付验证清单 -- [ ] 确认要查看的特定网络表(internet, route 等)。 -- [ ] 验证输出是否需要为自动化提供结构化格式 (CSV/TSV)。 diff --git a/data/x-cmd/node/SKILL.md b/data/x-cmd/node/SKILL.md deleted file mode 100644 index 3f0b98c..0000000 --- a/data/x-cmd/node/SKILL.md +++ /dev/null @@ -1,34 +0,0 @@ ---- -name: node -description: > - Enhanced Node.js module for version management, package installation (npm/npx), and environment setup. - Core Scenario: When the user needs to manage Node.js versions, run JS scripts, or install npm packages. -license: MIT ---- - -# node - Node.js Development Environment - -The `node` module simplifies Node.js development by offering integrated version management and direct access to npm and npx tools across various platforms. - -## When to Activate -- When installing or switching Node.js versions. -- When using `npm` to manage dependencies or `npx` to run remote tools. -- When executing JavaScript files in the terminal. - -## Patterns & Examples - -### Install NPM Packages -```bash -# Install a Node package globally -x node npm install -g typescript -``` - -### Run Tool via NPX -```bash -# Execute a tool without local installation -x node npx create-react-app my-app -``` - -## Checklist -- [ ] Verify the desired Node.js version. -- [ ] Ensure npm/npx is required for the specific task. diff --git a/data/x-cmd/node/SKILL.zh.md b/data/x-cmd/node/SKILL.zh.md deleted file mode 100644 index 0228f15..0000000 --- a/data/x-cmd/node/SKILL.zh.md +++ /dev/null @@ -1,34 +0,0 @@ ---- -name: node -description: > - 增强型 Node.js 模块,用于版本管理、包安装 (npm/npx) 和环境设置。 - 核心场景:当用户需要管理 Node.js 版本、运行 JS 脚本或安装 npm 包时。 -license: MIT ---- - -# node - Node.js 开发环境 - -`node` 模块通过提供集成的版本管理以及在各种平台上直接访问 npm 和 npx 工具,简化了 Node.js 开发。 - -## 激活时机 -- 当安装或切换 Node.js 版本时。 -- 当使用 `npm` 管理依赖项或使用 `npx` 运行远程工具时。 -- 当在终端中执行 JavaScript 文件时。 - -## 实战示例 - -### 安装 NPM 包 -```bash -# 全局安装一个 Node 包 -x node npm install -g typescript -``` - -### 通过 NPX 运行工具 -```bash -# 在不进行本地安装的情况下执行工具 -x node npx create-react-app my-app -``` - -## 交付验证清单 -- [ ] 验证所需的 Node.js 版本。 -- [ ] 确保特定任务需要 npm/npx。 diff --git a/data/x-cmd/npm/SKILL.md b/data/x-cmd/npm/SKILL.md deleted file mode 100644 index 99cec32..0000000 --- a/data/x-cmd/npm/SKILL.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: npm -description: > - Enhanced interface for npm, the Node.js package manager. - Core Scenario: When the user needs to manage JavaScript dependencies or run npm scripts. -license: MIT ---- - -# npm - Node.js Package Management - -The `npm` module provides integrated access to the Node Package Manager, simplifying dependency installation and script execution for Node.js projects. - -## When to Activate -- When installing or managing Node.js packages. -- When running predefined scripts in `package.json`. - -## Patterns & Examples - -### Install Global Package -```bash -# Install a package globally via npm -x npm install -g nodemon -``` - -## Checklist -- [ ] Confirm the package name. -- [ ] Verify the target project directory. diff --git a/data/x-cmd/npm/SKILL.zh.md b/data/x-cmd/npm/SKILL.zh.md deleted file mode 100644 index f3bdb17..0000000 --- a/data/x-cmd/npm/SKILL.zh.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: npm -description: > - npm 的增强接口,Node.js 软件包管理器。 - 核心场景:当用户需要管理 JavaScript 依赖项或运行 npm 脚本时。 -license: MIT ---- - -# npm - Node.js 包管理 - -`npm` 模块提供了对 Node 软件包管理器的集成访问,简化了 Node.js 项目的依赖项安装和脚本执行。 - -## 激活时机 -- 当安装或管理 Node.js 软件包时。 -- 当运行 `package.json` 中定义的脚本时。 - -## 实战示例 - -### 安装全局包 -```bash -# 通过 npm 全局安装一个包 -x npm install -g nodemon -``` - -## 交付验证清单 -- [ ] 确认包名。 -- [ ] 验证目标项目目录。 diff --git a/data/x-cmd/npx/SKILL.md b/data/x-cmd/npx/SKILL.md deleted file mode 100644 index 3f3a46a..0000000 --- a/data/x-cmd/npx/SKILL.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: npx -description: > - Enhanced interface for npx, used to execute Node.js packages without local installation. - Core Scenario: When the user needs to run a one-off Node.js tool or test a package. -license: MIT ---- - -# npx - Node Package Executor - -The `npx` module allows users to execute Node.js tools directly from the npm registry without needing to install them globally or locally first. - -## When to Activate -- When running a CLI tool once (e.g., scaffolding a project). -- When testing different versions of a package. - -## Patterns & Examples - -### Run Scaffolding Tool -```bash -# Execute create-react-app without installation -x npx create-react-app my-new-project -``` - -## Checklist -- [ ] Confirm the package name and arguments. -- [ ] Verify if a specific version is required. diff --git a/data/x-cmd/npx/SKILL.zh.md b/data/x-cmd/npx/SKILL.zh.md deleted file mode 100644 index 4c3b816..0000000 --- a/data/x-cmd/npx/SKILL.zh.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: npx -description: > - npx 的增强接口,用于在不进行本地安装的情况下执行 Node.js 软件包。 - 核心场景:当用户需要运行一次性 Node.js 工具或测试包时。 -license: MIT ---- - -# npx - Node 包执行器 - -`npx` 模块允许用户直接从 npm 注册表执行 Node.js 工具,而无需先全局或本地安装它们。 - -## 激活时机 -- 当运行一次性 CLI 工具时(如脚手架项目)。 -- 当测试包的不同版本时。 - -## 实战示例 - -### 运行脚手架工具 -```bash -# 在不安装的情况下执行 create-react-app -x npx create-react-app my-new-project -``` - -## 交付验证清单 -- [ ] 确认包名和参数。 -- [ ] 验证是否需要特定版本。 diff --git a/data/x-cmd/nu/SKILL.md b/data/x-cmd/nu/SKILL.md deleted file mode 100644 index 5846700..0000000 --- a/data/x-cmd/nu/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: nu -description: > - Enhanced interface for Nushell, enabling x-cmd integration, documentation search, and AI-powered command generation. - Core Scenario: When the user needs to set up x-cmd in Nushell or search Nushell documentation via CLI. -license: MIT ---- - -# nu - Nushell Enhancement & Integration - -The `nu` module improves the Nushell experience by providing easy integration with the x-cmd ecosystem and providing quick access to Nushell's official documentation and AI-generated commands. - -## When to Activate -- When the user wants to inject x-cmd tools (x, c, @gpt) into their Nushell environment. -- When searching for specific Nushell syntax or alias information on nushell.sh. -- When launching Nushell with automatic installation via x-cmd pkg. - -## Core Principles & Rules -- **Integration**: Use `--setup` to automatically modify `env.nu` for x-cmd support. -- **On-demand Setup**: Automatically handles Nushell installation if missing from the system. - -## Patterns & Examples - -### Setup x-cmd -```bash -# Inject x-cmd utilities into the Nushell configuration -x nu --setup -``` - -### Search Documentation -```bash -# Search nushell.sh for 'alias' information interactively -x nu : alias -``` - -## Checklist -- [ ] Confirm if the user is using Nushell as their primary shell. -- [ ] Verify if x-cmd should be injected permanently or just tested. diff --git a/data/x-cmd/nu/SKILL.zh.md b/data/x-cmd/nu/SKILL.zh.md deleted file mode 100644 index 4f72118..0000000 --- a/data/x-cmd/nu/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: nu -description: > - Nushell 的增强接口,支持 x-cmd 集成、文档搜索和 AI 驱动的命令生成。 - 核心场景:当用户需要在 Nushell 中设置 x-cmd 或通过 CLI 搜索 Nushell 文档时。 -license: MIT ---- - -# nu - Nushell 增强与集成 - -`nu` 模块通过提供与 x-cmd 生态系统的轻松集成,以及快速访问 Nushell 官方文档和 AI 生成的命令,提升了 Nushell 的体验。 - -## 激活时机 -- 当用户想要将 x-cmd 工具(x, c, @gpt)注入到其 Nushell 环境中时。 -- 当在 nushell.sh 网站上搜索特定的 Nushell 语法或别名信息时。 -- 当通过 x-cmd pkg 自动安装并启动 Nushell 时。 - -## 核心原则与规范 -- **集成**: 使用 `--setup` 自动修改 `env.nu` 以支持 x-cmd。 -- **按需安装**: 如果系统中缺少 Nushell,会自动处理其安装。 - -## 实战示例 - -### 设置 x-cmd -```bash -# 将 x-cmd 实用程序注入 Nushell 配置中 -x nu --setup -``` - -### 搜索文档 -```bash -# 在 nushell.sh 网站交互式搜索 'alias' 信息 -x nu : alias -``` - -## 交付验证清单 -- [ ] 确认用户是否将 Nushell 作为其主要 Shell。 -- [ ] 验证 x-cmd 应该是永久注入还是仅进行测试。 diff --git a/data/x-cmd/ollama/SKILL.md b/data/x-cmd/ollama/SKILL.md deleted file mode 100644 index 609726f..0000000 --- a/data/x-cmd/ollama/SKILL.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: ollama -description: > - CLI client for Ollama, the open-source framework for local LLM deployment. - Core Scenario: When the user wants to pull, manage, or run local AI models (Llama 3, Mistral, etc.) via Ollama. -license: MIT ---- - -# ollama - Local LLM Management & Interaction - -The `ollama` module provides an enhanced CLI interface for Ollama, enabling users to easily manage local models, perform translations, and interact with various AI architectures. - -## When to Activate -- When the user wants to install or manage Ollama services. -- When pulling models from the Ollama registry (e.g., Llama 3, Mistral). -- When chatting with local models or using them for file-based translation. -- When using an interactive UI to browse and download models. - -## Core Principles & Rules -- **Registry Interaction**: Use `pull` and `push` to sync models with the Ollama registry. -- **Convenience Aliases**: Support for `@o` alias for rapid local LLM chat. -- **File Context**: Leverage the `--file` flag to provide local context to models. - -## Additional Scenarios -- **Interactive Browsing**: Use `la` for an interactive UI to explore available models. -- **Service Control**: Use `serve` to manually start the Ollama backend. - -## Patterns & Examples - -### Pull and Run -```bash -# Download and start a chat session with Mistral -x ollama pull mistral -@o "How does a vector database work?" -``` - -### Translate with File Context -```bash -# Translate local documents using a local model -@o --file ./abstract.en.md "Translate to Chinese" -``` - -### Interactive UI -```bash -# Browse the Ollama registry interactively -x ollama la -``` - -## Checklist -- [ ] Ensure the Ollama service is installed and running. -- [ ] Verify if the desired model has been pulled to local storage. -- [ ] Check compatibility of the `--file` input with the model's context window. diff --git a/data/x-cmd/ollama/SKILL.zh.md b/data/x-cmd/ollama/SKILL.zh.md deleted file mode 100644 index 385fcc4..0000000 --- a/data/x-cmd/ollama/SKILL.zh.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: ollama -description: > - Ollama 命令行客户端,用于本地部署大语言模型的开源框架。 - 核心场景:当用户想要通过 Ollama 拉取、管理或运行本地 AI 模型(如 Llama 3, Mistral 等)时。 -license: MIT ---- - -# ollama - 本地大模型管理与交互 - -`ollama` 模块为 Ollama 提供了增强的命令行界面,使用户能够轻松管理本地模型、执行翻译并与各种 AI 架构进行交互。 - -## 激活时机 -- 当用户想要安装或管理 Ollama 服务时。 -- 当从 Ollama 注册表拉取模型(如 Llama 3, Mistral)时。 -- 当与本地模型聊天或将其用于基于文件的翻译时。 -- 当使用交互式 UI 浏览和下载模型时。 - -## 核心原则与规范 -- **注册表交互**: 使用 `pull` 和 `push` 与 Ollama 注册表同步模型。 -- **快捷别名**: 支持使用 `@o` 别名进行快速的本地模型对话。 -- **文件上下文**: 利用 `--file` 标志向模型提供本地上下文。 - -## 补充场景 -- **交互式浏览**: 使用 `la` 通过交互式 UI 探索可用模型。 -- **服务控制**: 使用 `serve` 手动启动 Ollama 后端。 - -## 实战示例 - -### 拉取并运行 -```bash -# 下载并开始与 Mistral 聊天 -x ollama pull mistral -@o "向量数据库是如何工作的?" -``` - -### 带有文件上下文的翻译 -```bash -# 使用本地模型翻译本地文档 -@o --file ./abstract.en.md "翻译为中文" -``` - -### 交互式 UI -```bash -# 以交互方式浏览 Ollama 注册表 -x ollama la -``` - -## 交付验证清单 -- [ ] 确保已安装并运行 Ollama 服务。 -- [ ] 验证所需的模型是否已拉取到本地存储。 -- [ ] 检查 `--file` 输入与模型上下文窗口的兼容性。 diff --git a/data/x-cmd/onsh/SKILL.md b/data/x-cmd/onsh/SKILL.md deleted file mode 100644 index c7405d3..0000000 --- a/data/x-cmd/onsh/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: onsh -description: > - Enhanced interface for xonsh, providing x-cmd integration, documentation search, and AI capabilities. - Core Scenario: When the user needs to integrate x-cmd with xonsh or search xonsh documentation via CLI. -license: MIT ---- - -# onsh - xonsh Python-Powered Shell Enhancement - -The `onsh` module enhances the xonsh shell experience, enabling a hybrid Python/Shell workflow integrated with the x-cmd toolset. - -## When to Activate -- When the user wants to inject x-cmd tools (x, c, @gpt) into the xonsh environment. -- When searching xon.sh for syntax, alias, or Python-integration tips. -- When launching xonsh with automatic setup via x-cmd pkg. - -## Core Principles & Rules -- **Integration**: Use `setup` to modify `.xonshrc` for permanent x-cmd support. -- **Xonsh Quirks**: Remind users that `@` aliases need a trailing `;` if no arguments are provided (e.g., `@gemini ;`). - -## Patterns & Examples - -### Setup xonsh -```bash -# Inject x-cmd utilities into the xonsh configuration -x onsh setup -``` - -### Search xonsh Docs -```bash -# Search xon.sh for 'alias' information interactively -x onsh : alias -``` - -## Checklist -- [ ] Verify if the user is aware of xonsh's Python-hybrid nature. -- [ ] Confirm if the `.xonshrc` file needs permanent modification. diff --git a/data/x-cmd/onsh/SKILL.zh.md b/data/x-cmd/onsh/SKILL.zh.md deleted file mode 100644 index accee17..0000000 --- a/data/x-cmd/onsh/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: onsh -description: > - xonsh 的增强接口,提供 x-cmd 集成、文档搜索和 AI 能力。 - 核心场景:当用户需要将 x-cmd 与 xonsh 集成或通过 CLI 搜索 xonsh 文档时。 -license: MIT ---- - -# onsh - xonsh Python 驱动 Shell 增强工具 - -`onsh` 模块增强了 xonsh Shell 的体验,实现了与 x-cmd 工具集集成的混合 Python/Shell 工作流。 - -## 激活时机 -- 当用户想要将 x-cmd 工具(x, c, @gpt)注入到 xonsh 环境中时。 -- 当在 xon.sh 网站上搜索语法、别名或 Python 集成提示时。 -- 当通过 x-cmd pkg 自动设置并启动 xonsh 时。 - -## 核心原则与规范 -- **集成**: 使用 `setup` 修改 `.xonshrc` 以实现永久的 x-cmd 支持。 -- **Xonsh 特性**: 提醒用户如果不提供参数,`@<名称>` 别名需要跟一个分号 `;`(例如 `@gemini ;`)。 - -## 实战示例 - -### 设置 xonsh -```bash -# 将 x-cmd 实用程序注入 xonsh 配置中 -x onsh setup -``` - -### 搜索 xonsh 文档 -```bash -# 在 xon.sh 网站交互式搜索 'alias' 信息 -x onsh : alias -``` - -## 交付验证清单 -- [ ] 验证用户是否知晓 xonsh 的 Python 混合特性。 -- [ ] 确认是否需要对 `.xonshrc` 文件进行永久修改。 diff --git a/data/x-cmd/openai/SKILL.md b/data/x-cmd/openai/SKILL.md deleted file mode 100644 index f457d27..0000000 --- a/data/x-cmd/openai/SKILL.md +++ /dev/null @@ -1,59 +0,0 @@ ---- -name: openai -description: > - Integrate OpenAI API for AI capabilities including ChatGPT conversations, text generation, image creation, and audio conversion. - Core Scenario: When the user needs to interact with OpenAI models for chat, translation, image generation, or audio processing. -license: MIT ---- - -# openai - OpenAI API Integration - -The `openai` module provides a comprehensive CLI interface for OpenAI services, enabling users to perform complex AI tasks like text generation, translation, image creation, and audio conversion directly from the terminal. - -## When to Activate -- When the user wants to use OpenAI for chat or text generation (e.g., generating commit messages). -- When the user needs to translate file contents using AI. -- When the user wants to generate images based on text prompts. -- When the user needs to convert text to speech or transcribe audio. -- When managing OpenAI API keys or model configurations. - -## Core Principles & Rules -- **API Key Management**: Use `init` or `--cfg apikey=` to set up the environment. -- **File Input**: Use the `--file` flag to provide context from local files to the chat/generation subcommands. -- **Reproducibility**: Be aware that AI outputs can vary; use appropriate parameters for consistent results if needed. - -## Additional Scenarios -- **Git Commit Messages**: Pipe `git diff` into `@gpt` to generate standardized commit messages. -- **Audio Processing**: Use the `audio` subcommand for TTS (Text-to-Speech) or transcription tasks. -- **Embedding & Fine-tuning**: Advanced users can manage fine-tuned models or calculate text embeddings. - -## Patterns & Examples - -### Chat with File Context -```bash -# Translate multiple files to Chinese using OpenAI -x openai chat request --file ./abstract.en.md --file ./content.en.md "Translate to chinese" -``` - -### Image Generation -```bash -# Generate an image based on a prompt -x openai image create --prompt "a high-quality digital art of a futuristic city" -``` - -### Text to Speech -```bash -# Convert text to an audio file -x openai audio generate --input "Welcome to x-cmd" --model tts-1 --voice alloy -``` - -### Generate Commit Message -```bash -# Pipe diff into gpt for commit message generation -git diff | @gpt "generate a suitable Git commit message that follows the Conventional Commits format" -``` - -## Checklist -- [ ] Ensure the OpenAI API key is configured using `x openai init`. -- [ ] Verify the input files exist when using the `--file` flag. -- [ ] Confirm the desired model and voice for audio generation. diff --git a/data/x-cmd/openai/SKILL.zh.md b/data/x-cmd/openai/SKILL.zh.md deleted file mode 100644 index 30db44f..0000000 --- a/data/x-cmd/openai/SKILL.zh.md +++ /dev/null @@ -1,59 +0,0 @@ ---- -name: openai -description: > - 集成 OpenAI API,提供包括 ChatGPT 对话、文本生成、图片创作及语音转换在内的 AI 能力。 - 核心场景:当用户需要与 OpenAI 模型交互进行聊天、翻译、图片生成或语音处理时。 -license: MIT ---- - -# openai - OpenAI API 集成 - -`openai` 模块提供 OpenAI 服务的完整命令行界面,使用户能够直接从终端执行复杂的 AI 任务,如文本生成、翻译、图片创建及语音转换。 - -## 激活时机 -- 当用户想要使用 OpenAI 进行聊天或文本生成时(例如,生成提交信息)。 -- 当用户需要使用 AI 翻译文件内容时。 -- 当用户想要根据文本提示生成图片时。 -- 当用户需要将文本转换为语音或转录音频时。 -- 当管理 OpenAI API 密钥或模型配置时。 - -## 核心原则与规范 -- **API 密钥管理**: 使用 `init` 或 `--cfg apikey=` 设置环境。 -- **文件输入**: 使用 `--file` 标志向聊天/生成子命令提供来自本地文件的上下文。 -- **重现性**: 请注意 AI 输出可能会有所不同;如果需要,请使用适当的参数以获得一致的结果。 - -## 补充场景 -- **Git 提交信息**: 将 `git diff` 传送到 `@gpt` 以生成标准化的提交信息。 -- **音频处理**: 将 `audio` 子命令用于 TTS(文本转语音)或转录任务。 -- **嵌入和微调**: 高级用户可以管理微调模型或计算文本嵌入。 - -## 实战示例 - -### 带有文件上下文的聊天 -```bash -# 使用 OpenAI 将多个文件翻译为中文 -x openai chat request --file ./abstract.en.md --file ./content.en.md "翻译为中文" -``` - -### 图片生成 -```bash -# 根据提示生成图片 -x openai image create --prompt "未来城市的的高质量数字艺术" -``` - -### 文本转语音 -```bash -# 将文本转换为音频文件 -x openai audio generate --input "欢迎来到 x-cmd" --model tts-1 --voice alloy -``` - -### 生成提交信息 -```bash -# 将 diff 传送到 gpt 以生成提交信息 -git diff | @gpt "生成符合 Conventional Commits 格式的合适 Git 提交信息" -``` - -## 交付验证清单 -- [ ] 确保已使用 `x openai init` 配置了 OpenAI API 密钥。 -- [ ] 使用 `--file` 标志时确认输入文件存在。 -- [ ] 确认语音生成的所需模型和声音。 diff --git a/data/x-cmd/openclaw/SKILL.md b/data/x-cmd/openclaw/SKILL.md deleted file mode 100644 index ae58887..0000000 --- a/data/x-cmd/openclaw/SKILL.md +++ /dev/null @@ -1,58 +0,0 @@ ---- -name: openclaw -description: > - Install and configure the OpenClaw personal AI assistant. - Core Scenario: When the user needs to deploy, uninstall, or configure their own local AI assistant on their device. -license: MIT ---- - -# openclaw - Personal AI Assistant Deployment - -The `openclaw` module provides a one-click initialization and configuration environment for OpenClaw, a personal AI assistant that can be run on local devices. - -## When to Activate -- When the user wants to install OpenClaw and its dependencies. -- When the user needs to set up specific integrations for OpenClaw (e.g., enterprise WeChat). -- When the user wants to manage the OpenClaw gateway service. -- When the user needs to uninstall OpenClaw and clean up its workspace. - -## Core Principles & Rules -- **Environment Automation**: Use `--install` to automatically detect and set up the necessary environment and dependencies. -- **Service Management**: Use the `service` subcommand to manage the persistent gateway process. -- **Integration Setup**: Use `--setup` to quickly configure models and chat software integrations. - -## Additional Scenarios -- **QYWX Integration**: Rapidly configure OpenClaw to work with Enterprise WeChat for communication. -- **Complete Cleanup**: Ensure all binaries, configs, and workspaces are removed during uninstallation. - -## Patterns & Examples - -### Install OpenClaw -```bash -# Install OpenClaw and its latest dependencies -x openclaw --install -``` - -### Setup QYWX Integration -```bash -# Configure OpenClaw to use Enterprise WeChat -x openclaw --setup qywx -``` - -### Manage Gateway Service -```bash -# Check status, start, or stop the gateway service -x openclaw service status -x openclaw service start -``` - -### Uninstall OpenClaw -```bash -# Remove all OpenClaw-related files and services -x openclaw --uninstall -``` - -## Checklist -- [ ] Verify that the user intends to perform an installation or configuration. -- [ ] Confirm if specific integrations (like `qywx`) are required during setup. -- [ ] Check the status of the gateway service if there are connectivity issues. diff --git a/data/x-cmd/openclaw/SKILL.zh.md b/data/x-cmd/openclaw/SKILL.zh.md deleted file mode 100644 index fc1e442..0000000 --- a/data/x-cmd/openclaw/SKILL.zh.md +++ /dev/null @@ -1,58 +0,0 @@ ---- -name: openclaw -description: > - 安装并配置 OpenClaw 个人 AI 助手。 - 核心场景:当用户需要在其设备上部署、卸载或配置自己的本地 AI 助手时。 -license: MIT ---- - -# openclaw - 个人 AI 助手部署 - -`openclaw` 模块为 OpenClaw(一款可以在个人设备上运行的 AI 助手)提供了一键式的环境初始化和配置。 - -## 激活时机 -- 当用户想要安装 OpenClaw 及其依赖项时。 -- 当用户需要为 OpenClaw 设置特定的集成(如企业微信)时。 -- 当用户想要管理 OpenClaw 的网关服务时。 -- 当用户需要卸载 OpenClaw 并清理其工作区时。 - -## 核心原则与规范 -- **环境自动化**: 使用 `--install` 自动检测并安装所需环境和依赖。 -- **服务管理**: 使用 `service` 子命令管理持久化的网关进程。 -- **配置集成**: 使用 `--setup` 快速配置模型和聊天软件集成。 - -## 补充场景 -- **企业微信集成**: 快速配置 OpenClaw 与企业微信进行通信。 -- **完整清理**: 确保在卸载过程中移除所有二进制文件、配置和工作区。 - -## 实战示例 - -### 安装 OpenClaw -```bash -# 安装 OpenClaw 及其最新版本的依赖环境 -x openclaw --install -``` - -### 配置企业微信集成 -```bash -# 配置 OpenClaw 接入企业微信 -x openclaw --setup qywx -``` - -### 管理网关服务 -```bash -# 查看状态、启动或停止网关服务 -x openclaw service status -x openclaw service start -``` - -### 卸载 OpenClaw -```bash -# 移除所有 OpenClaw 相关的二进制文件、配置文件和工作区 -x openclaw --uninstall -``` - -## 交付验证清单 -- [ ] 确认用户打算执行安装或配置操作。 -- [ ] 确认在设置过程中是否需要特定的集成(如 `qywx`)。 -- [ ] 如果存在连接问题,检查网关服务的状态。 diff --git a/data/x-cmd/openspec/SKILL.md b/data/x-cmd/openspec/SKILL.md deleted file mode 100644 index a1c492d..0000000 --- a/data/x-cmd/openspec/SKILL.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: openspec -description: > - Lightweight specification-driven framework to coordinate human and AI coding collaboration. - Core Scenario: When the user wants to initialize a project, manage change proposals, or validate project specs with AI. -license: MIT ---- - -# openspec - Spec-Driven Collaboration Framework - -The `openspec` module provides a CLI interface for OpenSpec, a framework designed to streamline human-AI collaboration through a standardized, specification-driven development process. - -## When to Activate -- When the user wants to initialize a new OpenSpec-driven project structure. -- When the user needs to manage change proposals (RFCs) or project specifications. -- When the user wants to validate consistency between project specifications and proposed changes. -- When the user needs an interactive dashboard to view the project's specs and progress. -- When updating the OpenSpec instruction files to better align AI assistants. - -## Core Principles & Rules -- **Spec-First**: Follow the flow of defining specifications (`spec`) before proposing changes (`change`). -- **Standardized Instructions**: Use `update` to ensure AI assistants have the latest alignment instructions. -- **Project Structure**: OpenSpec relies on specific `.md` files to track state; use `init` to set up the environment. - -## Additional Scenarios -- **Dashboard Overview**: Use `view` for an interactive summary of specs and changes. -- **Finalizing Changes**: Use `archive` to merge completed changes back into the main specifications. - -## Patterns & Examples - -### Initialize Project -```bash -# Set up OpenSpec structure in the current directory -x openspec init -``` - -### Create Change Proposal -```bash -# Start a new change proposal -x openspec change new -``` - -### Validate Alignment -```bash -# Check if current changes align with project specifications -x openspec validate -``` - -## Checklist -- [ ] Verify the project is initialized with `x openspec init`. -- [ ] Ensure AI assistant instructions are up-to-date by running `update`. diff --git a/data/x-cmd/openspec/SKILL.zh.md b/data/x-cmd/openspec/SKILL.zh.md deleted file mode 100644 index 12d497f..0000000 --- a/data/x-cmd/openspec/SKILL.zh.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: openspec -description: > - 轻量级规范驱动框架,用于协调人类与 AI 编码助理之间的协作。 - 核心场景:当用户想要初始化一个规范驱动的项目、管理变更提案或通过 AI 验证项目规范时。 -license: MIT ---- - -# openspec - 规范驱动的协作框架 - -`openspec` 模块为 OpenSpec 提供命令行界面,这是一款旨在通过标准化的、规范驱动开发流程来简化人机协作的框架。 - -## 激活时机 -- 当用户想要初始化一个新的 OpenSpec 驱动的项目结构时。 -- 当用户需要管理变更提案 (RFC) 或项目规范时。 -- 当用户想要验证项目规范与提议变更之间的一致性时。 -- 当用户需要一个交互式仪表板来查看项目规范和进度时。 -- 当更新 OpenSpec 指令文件以更好地与 AI 助手对齐时。 - -## 核心原则与规范 -- **规范先行**: 遵循在提议变更 (`change`) 之前定义规范 (`spec`) 的流程。 -- **标准化指令**: 使用 `update` 确保 AI 助手拥有最新的对齐指令。 -- **项目结构**: OpenSpec 依赖特定的 `.md` 文件来跟踪状态;使用 `init` 设置环境。 - -## 补充场景 -- **仪表板概览**: 使用 `view` 获取规范和变更的交互式摘要。 -- **完成变更**: 使用 `archive` 将已完成的变更合并回主规范。 - -## 实战示例 - -### 初始化项目 -```bash -# 在当前目录下设置 OpenSpec 结构 -x openspec init -``` - -### 创建变更提案 -```bash -# 开始一个新的变更提案 -x openspec change new -``` - -### 验证对齐 -```bash -# 检查当前变更是否与项目规范对齐 -x openspec validate -``` - -## 交付验证清单 -- [ ] 验证项目已使用 `x openspec init` 初始化。 -- [ ] 运行 `update` 确保 AI 助手的指令是最新的。 diff --git a/data/x-cmd/os/SKILL.md b/data/x-cmd/os/SKILL.md deleted file mode 100644 index d32b103..0000000 --- a/data/x-cmd/os/SKILL.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: os -description: > - Retrieve operating system information including architecture, release version, and uptime. - Core Scenario: When the user needs to check system specs, platform type, or runtime duration for scripting or diagnostics. -license: MIT ---- - -# os - Operating System Information - -The `os` module provides detailed information about the current operating system, including its name, architecture, release details, load averages, and more. It is essential for writing cross-platform scripts. - -## When to Activate -- When the user needs to identify the system architecture (e.g., x86_64, arm64). -- When checking if the environment is Linux, macOS, WSL, or Termux. -- When monitoring system load or uptime. -- When retrieving release information for software compatibility checks. - -## Core Principles & Rules -- **Efficient Scripting**: Use variables like `name_`, `arch_`, and `uptime_` in scripts for high-performance data retrieval. -- **Cache Control**: Subcommands starting with `get` (e.g., `getname`) bypass caches for real-time detection. -- **Cross-Platform Readiness**: Leverage `os is ` for robust conditional logic in shell scripts. - -## Patterns & Examples - -### Basic Information -```bash -# Show OS name and architecture -x os name -x os arch -``` - -### Environment Check -```bash -# Test if running in WSL (returns 0 or 1) -x os is wsl -``` - -### System Performance -```bash -# Display uptime and load averages -x os uptime -x os loadavg -``` - -## Checklist -- [ ] Confirm if real-time detection is required (bypassing cache). -- [ ] Verify the specific information requested (name vs release). diff --git a/data/x-cmd/os/SKILL.zh.md b/data/x-cmd/os/SKILL.zh.md deleted file mode 100644 index 7705da1..0000000 --- a/data/x-cmd/os/SKILL.zh.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: os -description: > - 获取操作系统信息,包括架构、发行版本和运行时间。 - 核心场景:当用户需要检查系统规格、平台类型或运行时长以进行脚本编写或诊断时。 -license: MIT ---- - -# os - 操作系统信息 - -`os` 模块提供有关当前操作系统的详细信息,包括其名称、架构、发行详情、平均负载等。这对于编写跨平台脚本至关重要。 - -## 激活时机 -- 当用户需要识别系统架构时(例如,x86_64, arm64)。 -- 当检查环境是否为 Linux、macOS、WSL 或 Termux 时。 -- 当监控系统负载或运行时间时。 -- 当获取发行版信息以进行软件兼容性检查时。 - -## 核心原则与规范 -- **高效脚本**: 在脚本中使用 `name_`、`arch_` 和 `uptime_` 等变量进行高性能数据检索。 -- **缓存控制**: 以 `get` 开头的子命令(如 `getname`)会绕过缓存进行实时检测。 -- **跨平台准备**: 利用 `os is <类型>` 在 Shell 脚本中实现健壮的条件逻辑。 - -## 实战示例 - -### 基本信息 -```bash -# 显示操作系统名称和架构 -x os name -x os arch -``` - -### 环境检查 -```bash -# 测试是否在 WSL 中运行(返回 0 或 1) -x os is wsl -``` - -### 系统性能 -```bash -# 显示运行时间和平均负载 -x os uptime -x os loadavg -``` - -## 交付验证清单 -- [ ] 确认是否需要实时检测(绕过缓存)。 -- [ ] 验证所请求的特定信息(名称与发行版)。 diff --git a/data/x-cmd/osv/SKILL.md b/data/x-cmd/osv/SKILL.md deleted file mode 100644 index 3f62b7a..0000000 --- a/data/x-cmd/osv/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: osv -description: > - CLI for Google's Open Source Vulnerabilities (OSV) project, providing vulnerability scanning and dependency analysis. - Core Scenario: When the user needs to scan projects for vulnerabilities or query specific CVE/OSV IDs. -license: MIT ---- - -# osv - Open Source Vulnerabilities Scanner - -The `osv` module provides an interface for the OSV project, enabling users to scan local projects, identify vulnerable dependencies, and retrieve detailed vulnerability information. - -## When to Activate -- When the user wants to perform a security audit on their project dependencies (npm, pip, etc.). -- When querying detailed information for a specific vulnerability ID (e.g., `osv-2020-111`). -- When generating security reports in the SARIF format for integration with CI/CD pipelines. -- When searching for vulnerabilities related to specific software packages and versions. - -## Core Principles & Rules -- **Comprehensive Scanning**: Use `sarif` to generate standardized security reports. -- **Eco-System Aware**: Supports multiple ecosystems including npm, pypi, and more. -- **Search Integration**: Uses AI or DuckDuckGo to summarize vulnerability details from the web. - -## Patterns & Examples - -### Full Project Scan -```bash -# Scan dependencies and generate a SARIF report -x osv sarif -``` - -### Specific Vulnerability Query -```bash -# Get details for a specific vulnerability ID -x osv vuln OSV-2020-111 -``` - -### Check Software Version -```bash -# Query vulnerabilities for a specific version of a package -x osv q -p jq -v 1.7.1 -``` - -## Checklist -- [ ] Confirm if the user needs a full scan or info on a specific ID. -- [ ] Verify the target project directory or package name. -- [ ] Ensure the correct ecosystem (pip, npm) is identified if using granular subcommands. diff --git a/data/x-cmd/osv/SKILL.zh.md b/data/x-cmd/osv/SKILL.zh.md deleted file mode 100644 index ceef050..0000000 --- a/data/x-cmd/osv/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: osv -description: > - Google 开源漏洞 (OSV) 项目的命令行工具,提供漏洞扫描和依赖分析。 - 核心场景:当用户需要扫描项目漏洞或查询特定的 CVE/OSV ID 时。 -license: MIT ---- - -# osv - 开源漏洞扫描器 - -`osv` 模块为 OSV 项目提供了界面,使用户能够扫描本地项目、识别有漏洞的依赖项并获取详细的漏洞信息。 - -## 激活时机 -- 当用户想要对其项目依赖项(npm, pip 等)执行安全审计时。 -- 当查询特定漏洞 ID(如 `osv-2020-111`)的详细信息时。 -- 当生成 SARIF 格式的安全报告以便与 CI/CD 流水线集成时。 -- 当搜索与特定软件包和版本相关的漏洞时。 - -## 核心原则与规范 -- **全面扫描**: 使用 `sarif` 生成标准化的安全报告。 -- **生态系统感知**: 支持包括 npm、pypi 在内的多个生态系统。 -- **搜索集成**: 利用 AI 或 DuckDuckGo 从网络总结漏洞详情。 - -## 实战示例 - -### 全项目扫描 -```bash -# 扫描依赖并生成 SARIF 报告 -x osv sarif -``` - -### 特定漏洞查询 -```bash -# 获取特定漏洞 ID 的详情 -x osv vuln OSV-2020-111 -``` - -### 检查软件版本 -```bash -# 查询特定版本软件包的漏洞 -x osv q -p jq -v 1.7.1 -``` - -## 交付验证清单 -- [ ] 确认用户需要全面扫描还是特定 ID 的信息。 -- [ ] 验证目标项目目录或软件包名称。 -- [ ] 如果使用细粒度的子命令,请确保识别了正确的生态系统(pip, npm)。 diff --git a/data/x-cmd/pandoc/SKILL.md b/data/x-cmd/pandoc/SKILL.md deleted file mode 100644 index f81f03a..0000000 --- a/data/x-cmd/pandoc/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: pandoc -description: > - Universal document converter supporting conversions between Markdown, HTML, PDF, and more. - Core Scenario: When the user needs to convert document formats or fetch web content as Markdown. -license: MIT ---- - -# pandoc - Universal Document Converter - -The `pandoc` module provides an interface for the powerful Pandoc document converter. It simplifies the installation process and enables easy conversion between diverse text and document formats. - -## When to Activate -- When converting Markdown files to HTML, PDF, or Word documents. -- When fetching and converting webpage content into clean Markdown. -- When listing supported document formats or extensions. -- When needing a zero-setup document conversion tool. - -## Core Principles & Rules -- **Zero-Setup**: Automatically downloads and manages the pandoc binary if it is missing. -- **Versatility**: Emphasize its ability to handle "anything to anything" conversions. -- **Standalone Support**: Encourage using the `-s` flag for full documents with headers/footers. - -## Patterns & Examples - -### Markdown to HTML -```bash -# Convert a Markdown file to a standalone HTML page -x pandoc -s input.md -o output.html -``` - -### Web to Markdown -```bash -# Convert a live webpage into a Markdown document -x pandoc -s -r html https://example.com -o webpage.md -``` - -### List Formats -```bash -# View all supported input document formats -x pandoc --list-input-formats -``` - -## Checklist -- [ ] Confirm the source and target document formats. -- [ ] Verify if a standalone document (`-s`) or a fragment is needed. -- [ ] Ensure input URLs or file paths are correct. diff --git a/data/x-cmd/pandoc/SKILL.zh.md b/data/x-cmd/pandoc/SKILL.zh.md deleted file mode 100644 index 139c1c3..0000000 --- a/data/x-cmd/pandoc/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: pandoc -description: > - 万能文档转换器,支持在 Markdown, HTML, PDF 等格式之间进行转换。 - 核心场景:当用户需要转换文档格式或将网页内容获取为 Markdown 时。 -license: MIT ---- - -# pandoc - 万能文档转换器 - -`pandoc` 模块为强大的 Pandoc 文档转换器提供了一个界面。它简化了安装过程,并能轻松地在多种文本和文档格式之间进行转换。 - -## 激活时机 -- 当将 Markdown 文件转换为 HTML、PDF 或 Word 文档时。 -- 当获取网页内容并将其转换为干净的 Markdown 时。 -- 当列出支持的文档格式或扩展名时。 -- 当需要一个零设置的文档转换工具时。 - -## 核心原则与规范 -- **零设置**: 如果二进制文件缺失,会自动下载并管理 pandoc。 -- **多功能性**: 强调其处理“任何格式到任何格式”转换的能力。 -- **独立支持**: 鼓励使用 `-s` 标志以获取带有页眉/页脚的完整文档。 - -## 实战示例 - -### Markdown 转 HTML -```bash -# 将 Markdown 文件转换为独立的 HTML 页面 -x pandoc -s input.md -o output.html -``` - -### 网页转 Markdown -```bash -# 将在线网页转换为 Markdown 文档 -x pandoc -s -r html https://example.com -o webpage.md -``` - -### 列出格式 -```bash -# 查看所有支持的输入文档格式 -x pandoc --list-input-formats -``` - -## 交付验证清单 -- [ ] 确认源文档和目标文档的格式。 -- [ ] 验证是需要独立文档 (`-s`) 还是代码片段。 -- [ ] 确保输入 URL 或文件路径正确。 diff --git a/data/x-cmd/path/SKILL.md b/data/x-cmd/path/SKILL.md deleted file mode 100644 index f3e2365..0000000 --- a/data/x-cmd/path/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: path -description: > - Manage the system PATH variable with subcommands for adding, removing, and deduplicating directories. - Core Scenario: When the user needs to modify their environment's search path or clean up duplicate entries. -license: MIT ---- - -# path - PATH Variable Management - -The `path` module provides a simple and effective way to manipulate the shell's PATH variable. It supports pushing/unshifting directories and deduplicating entries to maintain a clean environment. - -## When to Activate -- When the user wants to add a new directory to the start (`unshift`) or end (`push`) of the PATH. -- When removing specific directories from the PATH. -- When cleaning up duplicate directory entries in the PATH variable. -- When viewing the current PATH structure in an interactive app. - -## Core Principles & Rules -- **Non-Destructive**: Provides explicit subcommands for modification rather than direct string editing. -- **Interactive UI**: Use the default `x path` to browse the current PATH entries in a structured list. -- **Ordering**: Be mindful of whether a path should take priority (start of PATH) or be a fallback (end of PATH). - -## Patterns & Examples - -### Add to PATH -```bash -# Add a directory to the beginning of the PATH -x path unshift /usr/local/custom/bin -``` - -### Remove Duplicates -```bash -# Remove all redundant entries from the PATH variable -x path uniq -``` - -### View and List -```bash -# List all directories currently in the PATH -x path ls -``` - -## Checklist -- [ ] Confirm if the directory should be added to the front or back of the PATH. -- [ ] Verify the target directory path exists and is correct. diff --git a/data/x-cmd/path/SKILL.zh.md b/data/x-cmd/path/SKILL.zh.md deleted file mode 100644 index ea004b5..0000000 --- a/data/x-cmd/path/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: path -description: > - 管理系统 PATH 变量,具有添加、删除和去重目录的子命令。 - 核心场景:当用户需要修改其环境的搜索路径或清理重复条目时。 -license: MIT ---- - -# path - PATH 变量管理 - -`path` 模块提供了一种简单有效的方式来操作 Shell 的 PATH 变量。它支持在路径中添加/删除目录,并对条目进行去重以保持环境清洁。 - -## 激活时机 -- 当用户想要将新目录添加到 PATH 的开头 (`unshift`) 或末尾 (`push`) 时。 -- 当从 PATH 中删除特定目录时。 -- 当清理 PATH 变量中重复的目录条目时。 -- 当在交互式应用中查看当前 PATH 结构时。 - -## 核心原则与规范 -- **非破坏性**: 提供明确的子命令进行修改,而不是直接进行字符串编辑。 -- **交互式 UI**: 使用默认的 `x path` 在结构化列表中浏览当前 PATH 条目。 -- **优先级**: 注意路径是应该具有优先级(PATH 开头)还是作为回退(PATH 末尾)。 - -## 实战示例 - -### 添加到 PATH -```bash -# 将目录添加到 PATH 的开头 -x path unshift /usr/local/custom/bin -``` - -### 移除重复项 -```bash -# 移除 PATH 变量中所有多余的条目 -x path uniq -``` - -### 查看与列出 -```bash -# 列出 PATH 中当前引入的所有目录 -x path ls -``` - -## 交付验证清单 -- [ ] 确认目录应该添加到 PATH 的前面还是后面。 -- [ ] 验证目标目录路径是否存在且正确。 diff --git a/data/x-cmd/pb/SKILL.md b/data/x-cmd/pb/SKILL.md deleted file mode 100644 index 0e646e3..0000000 --- a/data/x-cmd/pb/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: pb -description: > - Cross-platform clipboard tool with support for local and remote (SSH) sessions. - Core Scenario: When the user needs to copy/paste text via terminal, including across SSH connections using OSC52. -license: MIT ---- - -# pb - Cross-Platform Clipboard - -The `pb` module provides a unified interface for system clipboards across macOS, Windows, Linux, and Android. It automatically detects the environment and even supports SSH-based copying through the OSC52 protocol. - -## When to Activate -- When the user wants to copy command output or text to the system clipboard. -- When pasting clipboard content into the terminal or a command. -- When working in a remote SSH session and needing to sync the remote clipboard with the local machine. - -## Core Principles & Rules -- **Environment Agnostic**: One command (`x pb`) replaces `pbcopy`, `xclip`, `wl-copy`, and `clip.exe`. -- **Remote Synchronization**: Automatically uses OSC52 in SSH sessions to allow copying from a server to a local client. -- **Input Flexibility**: Supports both piped input and direct arguments for copying. - -## Patterns & Examples - -### Copy from Pipe -```bash -# Copy the result of a command to clipboard -echo "data to copy" | x pb -``` - -### Copy Arguments -```bash -# Copy specific text directly -x pb copy "Hello World" -``` - -### Paste to Terminal -```bash -# Paste current clipboard content -x pb paste -``` - -## Checklist -- [ ] Confirm if the operation is a copy or a paste. -- [ ] Ensure the system has a clipboard manager installed if on Linux (xclip/wl-copy). -- [ ] Verify SSH client support for OSC52 if working remotely. diff --git a/data/x-cmd/pb/SKILL.zh.md b/data/x-cmd/pb/SKILL.zh.md deleted file mode 100644 index 1e32a77..0000000 --- a/data/x-cmd/pb/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: pb -description: > - 跨平台剪贴板工具,支持本地和远程 (SSH) 会话。 - 核心场景:当用户需要通过终端复制/粘贴文本时,包括使用 OSC52 在 SSH 连接中跨设备同步。 -license: MIT ---- - -# pb - 跨平台剪贴板工具 - -`pb` 模块为 macOS、Windows、Linux 和 Android 的系统剪贴板提供了统一的界面。它会自动检测环境,甚至支持通过 OSC52 协议在 SSH 会话中进行复制。 - -## 激活时机 -- 当用户想要将命令输出或文本复制到系统剪贴板时。 -- 当将剪贴板内容粘贴到终端或命令中时。 -- 当在远程 SSH 会话中工作,并需要将远程剪贴板与本地机器同步时。 - -## 核心原则与规范 -- **环境无关**: 一个命令 (`x pb`) 即可取代 `pbcopy`, `xclip`, `wl-copy` 和 `clip.exe`。 -- **远程同步**: 在 SSH 会话中自动使用 OSC52,允许将内容从服务器复制到本地客户端。 -- **输入灵活性**: 同时支持管道输入和直接参数输入进行复制。 - -## 实战示例 - -### 从管道复制 -```bash -# 将命令结果复制到剪贴板 -echo "要复制的数据" | x pb -``` - -### 复制参数内容 -```bash -# 直接复制特定文本 -x pb copy "你好世界" -``` - -### 粘贴到终端 -```bash -# 粘贴当前剪贴板内容 -x pb paste -``` - -## 交付验证清单 -- [ ] 确认操作是复制还是粘贴。 -- [ ] 如果在 Linux 上,确保系统已安装剪贴板管理器 (xclip/wl-copy)。 -- [ ] 如果远程工作,验证 SSH 客户端是否支持 OSC52。 diff --git a/data/x-cmd/perl/SKILL.md b/data/x-cmd/perl/SKILL.md deleted file mode 100644 index 9d5398c..0000000 --- a/data/x-cmd/perl/SKILL.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: perl -description: > - Enhanced Perl language module for script execution and environment management. - Core Scenario: When the user needs to run Perl scripts or manage the Perl runtime environment. -license: MIT ---- - -# perl - Perl Language Management - -The `perl` module provides a unified interface for executing Perl scripts and managing the Perl environment across various platforms. - -## When to Activate -- When running Perl scripts in the terminal. -- When identifying or managing the Perl runtime. - -## Patterns & Examples - -### Run Script -```bash -# Execute a local Perl script -x perl ./script.pl -``` - -## Checklist -- [ ] Verify the Perl script path. diff --git a/data/x-cmd/perl/SKILL.zh.md b/data/x-cmd/perl/SKILL.zh.md deleted file mode 100644 index 53fca00..0000000 --- a/data/x-cmd/perl/SKILL.zh.md +++ /dev/null @@ -1,26 +0,0 @@ ---- -name: perl -description: > - 增强型 Perl 语言模块,用于脚本执行和环境管理。 - 核心场景:当用户需要运行 Perl 脚本或管理 Perl 运行时环境时。 -license: MIT ---- - -# perl - Perl 语言管理 - -`perl` 模块为在各种平台上执行 Perl 脚本和管理 Perl 环境提供了一个统一的界面。 - -## 激活时机 -- 当在终端中运行 Perl 脚本时。 -- 当识别或管理 Perl 运行时。 - -## 实战示例 - -### 运行脚本 -```bash -# 执行本地 Perl 脚本 -x perl ./script.pl -``` - -## 交付验证清单 -- [ ] 验证 Perl 脚本路径。 diff --git a/data/x-cmd/pick/SKILL.md b/data/x-cmd/pick/SKILL.md deleted file mode 100644 index 331bb84..0000000 --- a/data/x-cmd/pick/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: pick -description: > - Interactive choice list generator for terminal, used to select items from stdin or command output. - Core Scenario: When the user needs to create an interactive selection menu for scripts or CLI workflows. -license: MIT ---- - -# pick - Interactive Selection Menu - -The `pick` module allows users to create interactive, searchable selection lists in the terminal. It is ideal for choosing files, processes, or custom options within a shell script. - -## When to Activate -- When a script needs to present a list of options for the user to select from. -- When choosing specific files from a large directory listing (`ls | x pick`). -- When performing multi-selection tasks within a terminal workflow (`--limit`). - -## Core Principles & Rules -- **Piping Input**: Automatically reads from `stdin` to generate the list. -- **Customization**: Supports setting the number of columns, rows, and selection limits. -- **Output**: Returns the selected item(s) to `stdout`, making it perfect for subshell execution. - -## Patterns & Examples - -### Select File -```bash -# Interactively select a file and view its status -stat `ls | x pick` -``` - -### Multi-Selection -```bash -# Select up to two items from a list using Tab -ls | x pick --limit 2 -``` - -### Styled Menu -```bash -# Create a 2-column selection menu with a custom prompt -ls | x pick --col 2 --ask "Choose a project:" -``` - -## Checklist -- [ ] Confirm the input data source (pipe or argument). -- [ ] Verify if single or multiple selection is required. -- [ ] Ensure the prompt message (`--ask`) is clear. diff --git a/data/x-cmd/pick/SKILL.zh.md b/data/x-cmd/pick/SKILL.zh.md deleted file mode 100644 index 15a08fb..0000000 --- a/data/x-cmd/pick/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: pick -description: > - 终端交互式选择列表生成器,用于从标准输入或命令输出中选择项目。 - 核心场景:当用户需要为脚本或 CLI 工作流创建交互式选择菜单时。 -license: MIT ---- - -# pick - 交互式选择菜单 - -`pick` 模块允许用户在终端中创建交互式、可搜索的选择列表。它是脚本中选择文件、进程或自定义选项的理想选择。 - -## 激活时机 -- 当脚本需要向用户呈现选项列表以供选择时。 -- 当从大型目录列表中选择特定文件时 (`ls | x pick`)。 -- 当在终端工作流中执行多选任务时 (`--limit`)。 - -## 核心原则与规范 -- **管道输入**: 自动从 `stdin` 读取以生成列表。 -- **自定义**: 支持设置列数、行数和选择限制。 -- **输出**: 将选择的项目返回到 `stdout`,非常适合子 shell 执行。 - -## 实战示例 - -### 选择文件 -```bash -# 交互式选择一个文件并查看其状态 -stat `ls | x pick` -``` - -### 多项选择 -```bash -# 使用 Tab 键从列表中最多选择两项 -ls | x pick --limit 2 -``` - -### 样式化菜单 -```bash -# 创建一个带有自定义提示的 2 列选择菜单 -ls | x pick --col 2 --ask "请选择一个项目:" -``` - -## 交付验证清单 -- [ ] 确认输入数据源(管道或参数)。 -- [ ] 验证是需要单选还是多选。 -- [ ] 确保提示信息 (`--ask`) 清晰。 diff --git a/data/x-cmd/ping/SKILL.md b/data/x-cmd/ping/SKILL.md deleted file mode 100644 index df2f69d..0000000 --- a/data/x-cmd/ping/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: ping -description: > - Enhanced ping command with visual data representations like heatmaps and bar charts. - Core Scenario: When the user needs to monitor network reachability with visual metrics. -license: MIT ---- - -# ping - Enhanced Network Reachability - -The `ping` module provides an enhanced interface for the standard connectivity test, supporting multiple visualization modes to better understand network performance and latency. - -## When to Activate -- When checking if a remote host is reachable via ICMP. -- When needing a visual representation (heatmap, bar chart) of network latency over time. -- When exporting reachability data to CSV or TSV for analysis. - -## Core Principles & Rules -- **Visualization**: Use `--heatmap` or `--bar` for immediate visual feedback on latency stability. -- **Data-Friendly**: Supports structured output formats for integration with monitoring tools. - -## Patterns & Examples - -### Visual Monitoring -```bash -# Ping a host and display a real-time heatmap of latency -x ping --heatmap bing.com -``` - -### Data Export -```bash -# Ping a host 10 times and output results as CSV -x ping -c 10 --csv 8.8.8.8 -``` - -## Checklist -- [ ] Confirm the target host address. -- [ ] Verify if a specific visualization mode is requested. diff --git a/data/x-cmd/ping/SKILL.zh.md b/data/x-cmd/ping/SKILL.zh.md deleted file mode 100644 index 0f57567..0000000 --- a/data/x-cmd/ping/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: ping -description: > - 增强型 ping 命令,具有热图和条形图等视觉数据表示。 - 核心场景:当用户需要使用视觉指标监控网络连通性时。 -license: MIT ---- - -# ping - 增强型网络连通性测试 - -`ping` 模块为标准的连通性测试提供了增强型界面,支持多种可视化模式,以更好地了解网络性能和延迟。 - -## 激活时机 -- 当检查远程主机是否通过 ICMP 可达时。 -- 当需要网络延迟随时间变化的视觉表示(热图、条形图)时。 -- 当为分析导出连通性数据到 CSV 或 TSV 时。 - -## 核心原则与规范 -- **可视化**: 使用 `--heatmap` 或 `--bar` 获取关于延迟稳定性的即时视觉反馈。 -- **数据友好**: 支持结构化输出格式,以便与监控工具集成。 - -## 实战示例 - -### 视觉监控 -```bash -# Ping 主机并显示延迟的实时热图 -x ping --heatmap bing.com -``` - -### 数据导出 -```bash -# Ping 主机 10 次并以 CSV 格式输出结果 -x ping -c 10 --csv 8.8.8.8 -``` - -## 交付验证清单 -- [ ] 确认目标主机地址。 -- [ ] 验证是否请求了特定的可视化模式。 diff --git a/data/x-cmd/pip/SKILL.md b/data/x-cmd/pip/SKILL.md deleted file mode 100644 index 265e3c3..0000000 --- a/data/x-cmd/pip/SKILL.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: pip -description: > - Enhanced interface for pip, the Python package installer. - Core Scenario: When the user needs to manage Python dependencies or search for packages on PyPI. -license: MIT ---- - -# pip - Python Package Management - -The `pip` module provides a convenient CLI for managing Python packages, ensuring that dependencies are handled correctly within the managed Python environment. - -## When to Activate -- When installing, upgrading, or removing Python packages. -- When listing currently installed Python dependencies. - -## Patterns & Examples - -### Install Package -```bash -# Install a specific Python package -x pip install pandas -``` - -## Checklist -- [ ] Confirm the package name. -- [ ] Verify the target Python environment. diff --git a/data/x-cmd/pip/SKILL.zh.md b/data/x-cmd/pip/SKILL.zh.md deleted file mode 100644 index 8411bfc..0000000 --- a/data/x-cmd/pip/SKILL.zh.md +++ /dev/null @@ -1,27 +0,0 @@ ---- -name: pip -description: > - pip 的增强接口,Python 软件包安装程序。 - 核心场景:当用户需要管理 Python 依赖项或在 PyPI 上搜索包时。 -license: MIT ---- - -# pip - Python 包管理 - -`pip` 模块为管理 Python 包提供了一个便捷的命令行界面,确保在受管 Python 环境中正确处理依赖项。 - -## 激活时机 -- 当安装、升级或移除 Python 软件包时。 -- 当列出当前安装的 Python 依赖项时。 - -## 实战示例 - -### 安装包 -```bash -# 安装特定的 Python 包 -x pip install pandas -``` - -## 交付验证清单 -- [ ] 确认包名。 -- [ ] 验证目标 Python 环境。 diff --git a/data/x-cmd/pixi/SKILL.md b/data/x-cmd/pixi/SKILL.md deleted file mode 100644 index 1331ea9..0000000 --- a/data/x-cmd/pixi/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: pixi -description: > - Cross-platform package manager and workflow tool based on Conda, with mirror acceleration. - Core Scenario: When the user needs fast, isolated environments for Python, C++, or other multi-language projects. -license: MIT ---- - -# pixi - Fast Package & Environment Manager - -The `pixi` module provides an enhanced interface for Pixi, offering mirror-accelerated downloads and simplified global package management using x-cmd conventions. - -## When to Activate -- When the user wants to install tools globally in their user environment (`use`). -- When managing conda-based projects with isolated dependencies. -- When creating a new project or adding dependencies to an existing pixi project. -- When running commands in a temporary, isolated environment (`exec`). - -## Core Principles & Rules -- **Global Management**: Prefer `x pixi use ` for user-wide tool installations. -- **Mirror Acceleration**: Automatically uses optimized mirrors for faster downloads in specific regions. -- **Environment Isolation**: Supports `shell` and `exec` for running code within a project's locked environment. - -## Patterns & Examples - -### Install Global Tools -```bash -# Install 'fish' shell and 'jq' globally via pixi -x pixi use fish jq -``` - -### Isolated Command Execution -```bash -# Run a command within a temporary pixi environment -x pixi exec python --version -``` - -### Interactive Browsing -```bash -# Interactively search for available pixi packages -x pixi la -``` - -## Checklist -- [ ] Confirm if the package should be installed globally or locally to a project. -- [ ] Verify if an isolated environment (`exec`) is preferred. -- [ ] Ensure the user is aware of the mirror acceleration benefits. diff --git a/data/x-cmd/pixi/SKILL.zh.md b/data/x-cmd/pixi/SKILL.zh.md deleted file mode 100644 index 502d2a1..0000000 --- a/data/x-cmd/pixi/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: pixi -description: > - 基于 Conda 的跨平台包管理器和工作流工具,具有镜像加速功能。 - 核心场景:当用户需要为 Python, C++ 或其他多语言项目提供快速、隔离的环境时。 -license: MIT ---- - -# pixi - 快速包与环境管理器 - -`pixi` 模块为 Pixi 提供了增强型界面,提供镜像加速下载,并使用 x-cmd 惯例简化了全局包管理。 - -## 激活时机 -- 当用户想要在其用户环境中全局安装工具时 (`use`)。 -- 当管理具有隔离依赖项的基于 Conda 的项目时。 -- 当创建一个新项目或向现有 pixi 项目添加依赖项时。 -- 当在临时的、隔离的环境中运行命令时 (`exec`)。 - -## 核心原则与规范 -- **全局管理**: 对于全系统工具安装,优先使用 `x pixi use `。 -- **镜像加速**: 自动使用优化的镜像以在特定地区加速下载。 -- **环境隔离**: 支持使用 `shell` 和 `exec` 在项目的锁定环境中运行代码。 - -## 实战示例 - -### 安装全局工具 -```bash -# 通过 pixi 全局安装 'fish' shell 和 'jq' -x pixi use fish jq -``` - -### 隔离的命令执行 -```bash -# 在临时的 pixi 环境中运行命令 -x pixi exec python --version -``` - -### 交互式浏览 -```bash -# 交互式搜索可用的 pixi 软件包 -x pixi la -``` - -## 交付验证清单 -- [ ] 确认软件包应该是全局安装还是局部安装到项目。 -- [ ] 验证是否首选隔离环境 (`exec`)。 -- [ ] 确保用户知晓镜像加速的好处。 diff --git a/data/x-cmd/pkgx/SKILL.md b/data/x-cmd/pkgx/SKILL.md deleted file mode 100644 index 3aaad4a..0000000 --- a/data/x-cmd/pkgx/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: pkgx -description: > - Universal tool runner for executing any open-source tool without pre-installation or config. - Core Scenario: When the user needs to run a CLI tool instantly without modifying their system or PATH. -license: MIT ---- - -# pkgx - Universal Tool Runner - -The `pkgx` module provides an interface for the pkgx tool runner, enabling users to fetch and run any open-source utility on-the-fly. It solves the "versioning" and "immediate availability" problems for developers. - -## When to Activate -- When the user wants to run a tool once without installing it globally. -- When testing a specific version of a software (e.g., `bun@1.0`). -- When adding temporary tools to the current execution environment. - -## Core Principles & Rules -- **Non-Invasive**: Runs tools without modifying the system's PATH or global configuration. -- **Immediate Execution**: Use `+pkg` to temporarily inject a tool into the environment for a specific command. -- **Zero-Setup**: Automatically handles tool acquisition and execution. - -## Patterns & Examples - -### Run Without Installation -```bash -# Execute bun without installing it to the system -x pkgx bun --eval 'console.log("hello world")' -``` - -### Temporary Environment Injection -```bash -# Temporarily add fzf to the current environment to run a command -eval "$(x pkgx +fzf)" -``` - -### Interactive Browsing -```bash -# Browse all available packages in the pkgx ecosystem -x pkgx ls -``` - -## Checklist -- [ ] Confirm if the tool is for one-off use or permanent installation. -- [ ] Verify if a specific version of the tool is required. -- [ ] Ensure the user is aware that the environment is not permanently modified. diff --git a/data/x-cmd/pkgx/SKILL.zh.md b/data/x-cmd/pkgx/SKILL.zh.md deleted file mode 100644 index d704680..0000000 --- a/data/x-cmd/pkgx/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: pkgx -description: > - 通用工具运行器,无需预先安装或配置即可执行任何开源工具。 - 核心场景:当用户需要立即运行 CLI 工具而不修改其系统或 PATH 时。 -license: MIT ---- - -# pkgx - 通用工具运行器 - -`pkgx` 模块为 pkgx 工具运行器提供了界面,使用户能够即时获取并运行任何开源实用程序。它为开发者解决了“版本控制”和“即时可用性”问题。 - -## 激活时机 -- 当用户想要运行一次工具而不想将其全局安装时。 -- 当测试软件的特定版本时(例如 `bun@1.0`)。 -- 当向当前执行环境添加临时工具时。 - -## 核心原则与规范 -- **非侵入性**: 在不修改系统 PATH 或全局配置的情况下运行工具。 -- **即时执行**: 使用 `+pkg` 将工具临时注入到特定命令的环境中。 -- **零设置**: 自动处理工具的获取和执行。 - -## 实战示例 - -### 无需安装运行 -```bash -# 在不向系统安装 bun 的情况下执行它 -x pkgx bun --eval 'console.log("hello world")' -``` - -### 临时环境注入 -```bash -# 在当前环境中临时添加 fzf 以执行命令 -eval "$(x pkgx +fzf)" -``` - -### 交互式浏览 -```bash -# 浏览 pkgx 生态系统中所有可用的软件包 -x pkgx ls -``` - -## 交付验证清单 -- [ ] 确认工具是用于一次性使用还是永久安装。 -- [ ] 验证是否需要该工具的特定版本。 -- [ ] 确保用户知晓环境不会被永久修改。 diff --git a/data/x-cmd/proxy/SKILL.md b/data/x-cmd/proxy/SKILL.md deleted file mode 100644 index e8e60ba..0000000 --- a/data/x-cmd/proxy/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: proxy -description: > - Manage shell environment proxy variables (http_proxy, etc.) with subcommands for setting, unsetting, and viewing. - Core Scenario: When the user needs to configure or clear network proxies for terminal tools and AI assistants. -license: MIT ---- - -# proxy - Shell Proxy Environment Management - -The `proxy` module simplifies the process of managing network proxies in the shell environment, providing a unified way to set or clear common proxy variables like `http_proxy` and `https_proxy`. - -## When to Activate -- When terminal tools or AI assistants face connectivity issues due to network restrictions. -- When setting up a temporary proxy for the current session. -- When clearing all proxy-related environment variables to restore direct connection. - -## Core Principles & Rules -- **Bulk Operations**: Setting a proxy targets all common proxy variables simultaneously. -- **Persistence**: Remind users that these changes typically apply only to the current shell session. -- **Clarity**: Use `ls` to audit which proxy variables are currently active. - -## Patterns & Examples - -### Set Proxy -```bash -# Point all proxy variables to a local address -x proxy set 127.0.0.1:7070 -``` - -### Clear Proxy -```bash -# Unset all proxy variables in the current session -x proxy unset -``` - -### View Active Proxies -```bash -# Audit active proxy-related environment variables -x proxy ls -``` - -## Checklist -- [ ] Confirm the target proxy address and port. -- [ ] Verify if the user intended to set or clear the proxy. -- [ ] Remind the user that this affects the current shell session. diff --git a/data/x-cmd/proxy/SKILL.zh.md b/data/x-cmd/proxy/SKILL.zh.md deleted file mode 100644 index f9af914..0000000 --- a/data/x-cmd/proxy/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: proxy -description: > - 管理 Shell 环境代理变量(http_proxy 等),具有设置、取消设置和查看的子命令。 - 核心场景:当用户需要为终端工具和 AI 助手配置或清除网络代理时。 -license: MIT ---- - -# proxy - Shell 代理环境管理 - -`proxy` 模块简化了在 Shell 环境中管理网络代理的过程,提供了一种统一的方法来设置或清除常见的代理变量,如 `http_proxy` 和 `https_proxy`。 - -## 激活时机 -- 当终端工具或 AI 助手由于网络限制而面临连接问题时。 -- 当为当前会话设置临时代理时。 -- 当清除所有与代理相关的环境变量以恢复直接连接时。 - -## 核心原则与规范 -- **批量操作**: 设置代理会同时针对所有常见的代理变量。 -- **持久性**: 提醒用户这些更改通常仅适用于当前的 Shell 会话。 -- **清晰性**: 使用 `ls` 审核当前激活的代理变量。 - -## 实战示例 - -### 设置代理 -```bash -# 将所有代理变量指向本地地址 -x proxy set 127.0.0.1:7070 -``` - -### 清除代理 -```bash -# 取消设置当前会话中的所有代理变量 -x proxy unset -``` - -### 查看激活的代理 -```bash -# 审核当前激活的与代理相关的环境变量 -x proxy ls -``` - -## 交付验证清单 -- [ ] 确认目标的代理地址和端口。 -- [ ] 验证用户是打算设置还是清除代理。 -- [ ] 提醒用户这会影响当前的 Shell 会话。 diff --git a/data/x-cmd/ps/SKILL.md b/data/x-cmd/ps/SKILL.md deleted file mode 100644 index 67cb4af..0000000 --- a/data/x-cmd/ps/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: ps -description: > - Display running process information with support for interactive UI and structured exports (CSV, JSON, TSV). - Core Scenario: When the user needs to analyze system processes or export process data for external automation. -license: MIT ---- - -# ps - Enhanced Process Status - -The `ps` module extends the standard process viewing capability by adding interactive TUI features and the ability to export process snapshots in structured data formats like JSON and CSV. - -## When to Activate -- When the user wants to browse processes using an interactive UI (FZF or CSV app). -- When exporting current process lists to JSON, CSV, or TSV for scripting. -- When converting legacy `ps aux` output into structured data formats. - -## Core Principles & Rules -- **Interactive Exploration**: Use `fz` for a fast, searchable TUI experience. -- **Data Exporting**: Prioritize `--json` or `--csv` when the user intends to process the data with other tools (like `jq`). -- **Piping Compatibility**: Can consume output from standard `ps` commands and transform it. - -## Patterns & Examples - -### Interactive UI -```bash -# View processes in a searchable interactive TUI -x ps fz -``` - -### Export to JSON -```bash -# Output current process information as a JSON array -x ps --json -``` - -### Data Transformation -```bash -# Convert a specific ps command output to CSV -ps -ef | x ps --tocsv -``` - -## Checklist -- [ ] Confirm if the user needs an interactive view or a static data export. -- [ ] Verify the desired output format (JSON, CSV, TSV). -- [ ] Ensure specific `ps` flags are used if narrowing down the process list. diff --git a/data/x-cmd/ps/SKILL.zh.md b/data/x-cmd/ps/SKILL.zh.md deleted file mode 100644 index b35e92c..0000000 --- a/data/x-cmd/ps/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: ps -description: > - 显示运行中的进程信息,支持交互式 UI 和结构化导出 (CSV, JSON, TSV)。 - 核心场景:当用户需要分析系统进程或为外部自动化导出进程数据时。 -license: MIT ---- - -# ps - 增强型进程状态查看 - -`ps` 模块通过添加交互式 TUI 功能以及将进程快照导出为 JSON 和 CSV 等结构化数据格式的能力,扩展了标准的进程查看功能。 - -## 激活时机 -- 当用户想要使用交互式 UI(FZF 或 CSV app)浏览进程时。 -- 当为脚本编写将当前进程列表导出为 JSON、CSV 或 TSV 时。 -- 当将传统的 `ps aux` 输出转换为结构化数据格式时。 - -## 核心原则与规范 -- **交互式探索**: 使用 `fz` 获得快速、可搜索的 TUI 体验。 -- **数据导出**: 当用户打算使用其他工具(如 `jq`)处理数据时,优先使用 `--json` 或 `--csv`。 -- **管道兼容性**: 可以接收标准 `ps` 命令的输出并对其进行转换。 - -## 实战示例 - -### 交互式 UI -```bash -# 在可搜索的交互式 TUI 中查看进程 -x ps fz -``` - -### 导出为 JSON -```bash -# 将当前进程信息输出为 JSON 数组 -x ps --json -``` - -### 数据转换 -```bash -# 将特定的 ps 命令输出转换为 CSV -ps -ef | x ps --tocsv -``` - -## 交付验证清单 -- [ ] 确认用户需要的是交互式视图还是静态数据导出。 -- [ ] 验证所需的输出格式(JSON, CSV, TSV)。 -- [ ] 如果缩小进程列表范围,确保使用了特定的 `ps` 标志。 diff --git a/data/x-cmd/pwsh/SKILL.md b/data/x-cmd/pwsh/SKILL.md deleted file mode 100644 index 1ce350e..0000000 --- a/data/x-cmd/pwsh/SKILL.md +++ /dev/null @@ -1,34 +0,0 @@ ---- -name: pwsh -description: > - PowerShell enhancement module providing interactive REPL and system management tools. - Core Scenario: When the user needs to manage system resources (disk, IP, processes) or use PowerShell via x-cmd. -license: MIT ---- - -# pwsh - PowerShell CLI Enhancement - -The `pwsh` module improves the PowerShell experience on supported platforms, offering an interactive REPL and powerful system management subcommands for disks, IPs, and processes. - -## When to Activate -- When the user needs to manage Windows/PowerShell-specific system resources. -- When entering a PowerShell interactive REPL via x-cmd. -- When performing rapid system diagnostics (processes, services, logs) using PowerShell. - -## Patterns & Examples - -### Interactive REPL -```bash -# Enter the interactive PowerShell REPL -x pwsh --repl -``` - -### System Management -```bash -# List system processes using PowerShell backend -x pwsh ps -``` - -## Checklist -- [ ] Confirm if the environment supports PowerShell (primarily for Git Bash/Windows). -- [ ] Verify the specific system resource being managed. diff --git a/data/x-cmd/pwsh/SKILL.zh.md b/data/x-cmd/pwsh/SKILL.zh.md deleted file mode 100644 index 7d10578..0000000 --- a/data/x-cmd/pwsh/SKILL.zh.md +++ /dev/null @@ -1,34 +0,0 @@ ---- -name: pwsh -description: > - PowerShell 增强模块,提供交互式 REPL 和系统管理工具。 - 核心场景:当用户需要管理系统资源(磁盘、IP、进程)或通过 x-cmd 使用 PowerShell 时。 -license: MIT ---- - -# pwsh - PowerShell 命令行增强工具 - -`pwsh` 模块在受支持的平台上提升了 PowerShell 的体验,提供了一个交互式 REPL 以及用于磁盘、IP 和进程的强大系统管理子命令。 - -## 激活时机 -- 当用户需要管理 Windows/PowerShell 特定的系统资源时。 -- 当通过 x-cmd 进入 PowerShell 交互式 REPL 时。 -- 当使用 PowerShell 执行快速系统诊断(进程、服务、日志)时。 - -## 实战示例 - -### 交互式 REPL -```bash -# 进入交互式 PowerShell REPL -x pwsh --repl -``` - -### 系统管理 -```bash -# 使用 PowerShell 后端列出系统进程 -x pwsh ps -``` - -## 交付验证清单 -- [ ] 确认环境是否支持 PowerShell(主要用于 Git Bash/Windows)。 -- [ ] 验证正在管理的特定系统资源。 diff --git a/data/x-cmd/python/SKILL.md b/data/x-cmd/python/SKILL.md deleted file mode 100644 index f8f860b..0000000 --- a/data/x-cmd/python/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: python -description: > - Enhanced Python environment manager, providing version management, pip integration, and automated setup. - Core Scenario: When the user needs to install specific Python versions, manage packages, or run scripts. -license: MIT ---- - -# python - Python Environment Management - -The `python` module provides a unified CLI for managing Python versions and environments, integrating with x-cmd's package system for seamless setup across diverse OS platforms. - -## When to Activate -- When installing or switching between Python versions. -- When managing Python packages using the integrated `pip` functionality. -- When running Python scripts or entering interactive REPLs. - -## Core Principles & Rules -- **Integration**: Works with asdf and other version managers to ensure availability. -- **Submodule Support**: Use `pip` for managing dependencies within the current Python environment. - -## Patterns & Examples - -### Install Packages -```bash -# Install a package globally within the managed Python environment -x python pip install requests -``` - -### Run Script -```bash -# Execute a Python script with the managed runtime -x python ./script.py -``` - -## Checklist -- [ ] Confirm the target Python version if multiple are available. -- [ ] Verify the package name for pip installation. diff --git a/data/x-cmd/python/SKILL.zh.md b/data/x-cmd/python/SKILL.zh.md deleted file mode 100644 index ec20ac6..0000000 --- a/data/x-cmd/python/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: python -description: > - 增强型 Python 环境管理器,提供版本管理、pip 集成和自动设置。 - 核心场景:当用户需要安装特定的 Python 版本、管理包或运行脚本时。 -license: MIT ---- - -# python - Python 环境管理 - -`python` 模块提供了一个统一的命令行界面来管理 Python 版本和环境,并与 x-cmd 的包系统集成,以便在不同的操作系统平台上实现无缝设置。 - -## 激活时机 -- 当安装或在不同的 Python 版本之间切换时。 -- 当使用集成的 `pip` 功能管理 Python 包时。 -- 当运行 Python 脚本或进入交互式 REPL 时。 - -## 核心原则与规范 -- **集成性**: 与 asdf 及其他版本管理器配合工作,以确保可用性。 -- **子模块支持**: 使用 `pip` 管理当前 Python 环境中的依赖项。 - -## 实战示例 - -### 安装包 -```bash -# 在受管 Python 环境中全局安装一个包 -x python pip install requests -``` - -### 运行脚本 -```bash -# 使用受管运行时执行 Python 脚本 -x python ./script.py -``` - -## 交付验证清单 -- [ ] 如果有多个可用版本,确认目标 Python 版本。 -- [ ] 验证用于 pip 安装的包名。 diff --git a/data/x-cmd/qywx/SKILL.md b/data/x-cmd/qywx/SKILL.md deleted file mode 100644 index 8c12e8e..0000000 --- a/data/x-cmd/qywx/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: qywx -description: > - Send messages to Enterprise WeChat (QYWX) groups using bot webhooks. - Core Scenario: When the user needs to automate notifications or send reports to a QYWX group. -license: MIT ---- - -# qywx - Enterprise WeChat Bot Utility - -The `qywx` module allows users to send various types of messages (text, markdown, images) to Enterprise WeChat groups via a configured bot webhook. - -## When to Activate -- When automating CI/CD notifications to a QYWX group. -- When sending system alerts or periodic reports via markdown. -- When uploading and sharing images within a QYWX group from the terminal. - -## Core Principles & Rules -- **Webhook Required**: Remind users to configure the `webhook` url via `init` or `--cfg`. -- **Message Types**: Supports `text`, `markdown`, and `image` formats. -- **Privacy**: Encourage keeping the webhook URL private. - -## Patterns & Examples - -### Send Markdown -```bash -# Send a formatted markdown notification -x qywx send --markdown "## Build Success\nVersion: v1.0.0" -``` - -### Send Image -```bash -# Upload and send a local image file -x qywx send --image ./report.png -``` - -### Configure Webhook -```bash -# Set up the bot webhook URL -x qywx --cfg webhook=YOUR_WEBHOOK_URL -``` - -## Checklist -- [ ] Ensure the webhook URL is correctly configured. -- [ ] Confirm the message type (text/markdown/image) matches the content. -- [ ] Verify file paths if sending an image. diff --git a/data/x-cmd/qywx/SKILL.zh.md b/data/x-cmd/qywx/SKILL.zh.md deleted file mode 100644 index 4b10797..0000000 --- a/data/x-cmd/qywx/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: qywx -description: > - 使用机器人 Webhook 向企业微信 (QYWX) 群组发送消息。 - 核心场景:当用户需要自动化通知或向企业微信群组发送报告时。 -license: MIT ---- - -# qywx - 企业微信机器人工具 - -`qywx` 模块允许用户通过配置好的机器人 Webhook 向企业微信群组发送各种类型的消息(文本、Markdown、图片)。 - -## 激活时机 -- 当向企业微信群组自动化发送 CI/CD 通知时。 -- 当通过 Markdown 发送系统警报或定期报告时。 -- 当从终端在企业微信群组内上传并共享图片时。 - -## 核心原则与规范 -- **需要 Webhook**: 提醒用户通过 `init` 或 `--cfg` 配置 `webhook` URL。 -- **消息类型**: 支持 `text`、`markdown` 和 `image` 格式。 -- **隐私**: 鼓励保持 Webhook URL 的私密性。 - -## 实战示例 - -### 发送 Markdown -```bash -# 发送格式化的 Markdown 通知 -x qywx send --markdown "## 构建成功\n版本:v1.0.0" -``` - -### 发送图片 -```bash -# 上传并发送本地图片文件 -x qywx send --image ./report.png -``` - -### 配置 Webhook -```bash -# 设置机器人 Webhook URL -x qywx --cfg webhook=你的_WEBHOOK_URL -``` - -## 交付验证清单 -- [ ] 确保 Webhook URL 已正确配置。 -- [ ] 确认消息类型(文本/Markdown/图片)与内容匹配。 -- [ ] 如果发送图片,请验证文件路径。 diff --git a/data/x-cmd/raku/SKILL.md b/data/x-cmd/raku/SKILL.md deleted file mode 100644 index 05682b8..0000000 --- a/data/x-cmd/raku/SKILL.md +++ /dev/null @@ -1,25 +0,0 @@ ---- -name: raku -description: > - Enhanced Raku language module for executing Raku (formerly Perl 6) scripts. - Core Scenario: When the user needs to run Raku scripts or manage the Raku environment. -license: MIT ---- - -# raku - Raku Language Management - -The `raku` module provides an interface for the Raku programming language, simplifying script execution and environment setup. - -## When to Activate -- When running Raku scripts or interacting with the Raku ecosystem. - -## Patterns & Examples - -### Run Raku Script -```bash -# Execute a Raku file -x raku ./myscript.raku -``` - -## Checklist -- [ ] Verify the Raku runtime is available. diff --git a/data/x-cmd/raku/SKILL.zh.md b/data/x-cmd/raku/SKILL.zh.md deleted file mode 100644 index d13431d..0000000 --- a/data/x-cmd/raku/SKILL.zh.md +++ /dev/null @@ -1,25 +0,0 @@ ---- -name: raku -description: > - 增强型 Raku 语言模块,用于执行 Raku(原名 Perl 6)脚本。 - 核心场景:当用户需要运行 Raku 脚本或管理 Raku 环境时。 -license: MIT ---- - -# raku - Raku 语言管理 - -`raku` 模块为 Raku 编程语言提供了界面,简化了脚本执行和环境设置。 - -## 激活时机 -- 当运行 Raku 脚本或与 Raku 生态系统交互时。 - -## 实战示例 - -### 运行 Raku 脚本 -```bash -# 执行一个 Raku 文件 -x raku ./myscript.raku -``` - -## 交付验证清单 -- [ ] 验证 Raku 运行时是否可用。 diff --git a/data/x-cmd/rand/SKILL.md b/data/x-cmd/rand/SKILL.md deleted file mode 100644 index 3eec073..0000000 --- a/data/x-cmd/rand/SKILL.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: rand -description: > - Random data generator for identifiers, numbers, strings, and mock data. - Core Scenario: When the user needs to generate UUIDs, random strings, or test data (emails, IPs) for scripting. -license: MIT ---- - -# rand - Random Data Generation - -The `rand` module generates various types of random data, ranging from basic numbers and strings to complex identifiers like UUIDv7 and mock data like emails or IP addresses. - -## When to Activate -- When generating unique identifiers (UUID, UUIDv7). -- When creating random passwords or tokens (strings, alphanum). -- When generating mock data for testing (emails, IPs). -- When requiring random integers or floats within a specific range. - -## Core Principles & Rules -- **Range Control**: Use specific min/max values for `int` and `float` to suit the task. -- **Bulk Generation**: Support for generating multiple items simultaneously (e.g., `x rand email 5`). -- **Identifier Uniqueness**: Prefer `uuidv7` for time-ordered unique identifiers. - -## Patterns & Examples - -### Generate Identifiers -```bash -# Create a standard UUID and a UUIDv7 -x rand uuid -x rand uuidv7 -``` - -### Random Strings -```bash -# Generate a 16-character random alphanumeric string -x rand alphanum 16 -``` - -### Mock Testing Data -```bash -# Generate 10 random IPv4 addresses -x rand ip 10 -``` - -## Checklist -- [ ] Confirm the required length or range for the random data. -- [ ] Verify if multiple results are needed. -- [ ] Ensure the correct data type (alpha, numeric, email, etc.) is selected. diff --git a/data/x-cmd/rand/SKILL.zh.md b/data/x-cmd/rand/SKILL.zh.md deleted file mode 100644 index 96ca1dd..0000000 --- a/data/x-cmd/rand/SKILL.zh.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: rand -description: > - 用于标识符、数字、字符串和模拟数据的随机数据生成器。 - 核心场景:当用户需要为脚本生成 UUID、随机字符串或测试数据(邮箱、IP)时。 -license: MIT ---- - -# rand - 随机数据生成 - -`rand` 模块生成各种类型的随机数据,从基础数字和字符串到 UUIDv7 等复杂标识符,以及邮箱或 IP 地址等模拟数据。 - -## 激活时机 -- 当生成唯一标识符(UUID, UUIDv7)时。 -- 当创建随机密码或令牌(字符串、字母数字)时。 -- 当为测试生成模拟数据(邮箱、IP)时。 -- 当需要特定范围内的随机整数或浮点数时。 - -## 核心原则与规范 -- **范围控制**: 为 `int` 和 `float` 使用特定的最小值/最大值以适应任务。 -- **批量生成**: 支持同时生成多个项目(例如 `x rand email 5`)。 -- **标识符唯一性**: 优先使用 `uuidv7` 以获得按时间排序的唯一标识符。 - -## 实战示例 - -### 生成标识符 -```bash -# 创建标准 UUID 和 UUIDv7 -x rand uuid -x rand uuidv7 -``` - -### 随机字符串 -```bash -# 生成 16 位随机字母数字字符串 -x rand alphanum 16 -``` - -### 模拟测试数据 -```bash -# 生成 10 个随机 IPv4 地址 -x rand ip 10 -``` - -## 交付验证清单 -- [ ] 确认随机数据所需的长度或范围。 -- [ ] 验证是否需要多个结果。 -- [ ] 确保选择了正确的数据类型(字母、数字、邮箱等)。 diff --git a/data/x-cmd/rg/SKILL.md b/data/x-cmd/rg/SKILL.md deleted file mode 100644 index c82b1e2..0000000 --- a/data/x-cmd/rg/SKILL.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: rg -description: > - Modern, high-speed line-oriented search tool (ripgrep) with an interactive FZF interface. - Core Scenario: When the user needs to perform recursive text searches in large codebases or directories. -license: MIT ---- - -# rg - High-Performance Text Search - -The `rg` module provides a powerful CLI for ripgrep, one of the fastest search tools available. It enhances the experience by adding an interactive FZF application for real-time searching and filtering. - -## When to Activate -- When the user needs to search for text patterns recursively across directories. -- When an interactive, searchable TUI is required for exploring code search results. -- When searching within compressed files (`-z`). -- When performing multi-line regex matching. - -## Core Principles & Rules -- **High Performance**: Emphasize that it handles large datasets efficiently. -- **Interactive Mode**: Use the default `x rg` (or `--fzfapp`) for a dynamic search experience. -- **Convenient Shortcuts**: In interactive mode, use `alt-a` to select all or `ctrl-o` to edit matches. - -## Additional Scenarios -- **File Type Filtering**: Use `-t` to limit search to specific file types (e.g., `python`, `js`). -- **Structured Data**: Supports `--json` for programmatic result processing. - -## Patterns & Examples - -### Interactive Search -```bash -# Start an interactive search in the current directory -x rg -``` - -### Search Specific Directory -```bash -# Open interactive TUI to search inside a specific path -x rg --fzfapp ~/.x-cmd.root -``` - -### Search Compressed Files -```bash -# Recursively search for patterns inside .zip or .gz files -x rg -z "search pattern" -``` - -## Checklist -- [ ] Confirm the search pattern or regex. -- [ ] Verify if an interactive or standard output is preferred. -- [ ] Check if the search should be limited to specific file types or depths. diff --git a/data/x-cmd/rg/SKILL.zh.md b/data/x-cmd/rg/SKILL.zh.md deleted file mode 100644 index b844df3..0000000 --- a/data/x-cmd/rg/SKILL.zh.md +++ /dev/null @@ -1,51 +0,0 @@ ---- -name: rg -description: > - 现代化的、高速的面问行的搜索工具 (ripgrep),具有交互式 FZF 界面。 - 核心场景:当用户需要在大型代码库或目录中执行递归文本搜索时。 -license: MIT ---- - -# rg - 高性能文本搜索 - -`rg` 模块为 ripgrep(最快的搜索工具之一)提供了强大的命令行界面。它通过添加 FZF 交互式应用来增强体验,实现实时的搜索和过滤。 - -## 激活时机 -- 当用户需要在目录中递归搜索文本模式时。 -- 当需要交互式的、可搜索的 TUI 来探索代码搜索结果时。 -- 当在压缩文件中搜索时 (`-z`)。 -- 当执行跨多行的正则表达式匹配时。 - -## 核心原则与规范 -- **高性能**: 强调它能高效地处理大型数据集。 -- **交互模式**: 使用默认的 `x rg`(或 `--fzfapp`)获得动态的搜索体验。 -- **便捷快捷键**: 在交互模式下,使用 `alt-a` 全选或 `ctrl-o` 编辑匹配项。 - -## 补充场景 -- **文件类型过滤**: 使用 `-t` 将搜索限制在特定的文件类型(如 `python`, `js`)。 -- **结构化数据**: 支持 `--json` 以进行编程化的结果处理。 - -## 实战示例 - -### 交互式搜索 -```bash -# 在当前目录启动交互式搜索 -x rg -``` - -### 搜索特定目录 -```bash -# 打开交互式 TUI 在特定路径内搜索 -x rg --fzfapp ~/.x-cmd.root -``` - -### 搜索压缩文件 -```bash -# 在 .zip 或 .gz 文件内递归搜索模式 -x rg -z "搜索模式" -``` - -## 交付验证清单 -- [ ] 确认搜索模式或正则表达式。 -- [ ] 验证首选的是交互式还是标准输出。 -- [ ] 检查搜索是否应限制在特定的文件类型或深度。 diff --git a/data/x-cmd/route/SKILL.md b/data/x-cmd/route/SKILL.md deleted file mode 100644 index e7da06a..0000000 --- a/data/x-cmd/route/SKILL.md +++ /dev/null @@ -1,37 +0,0 @@ ---- -name: route -description: > - Display the system's network routing table with support for structured data export. - Core Scenario: When the user needs to inspect the routing table for network path diagnostics. -license: MIT ---- - -# route - Network Routing Table Viewer - -The `route` module provides an interface to view the system's network routing table, allowing users to inspect how network traffic is directed. It supports exporting the table in structured formats. - -## When to Activate -- When diagnosing network reachability or routing issues. -- When needing a structured list (CSV) of the current routing table for analysis. - -## Core Principles & Rules -- **Experimental Status**: Be aware that this module is currently in an experimental phase. -- **Data Export**: Supports `--csv` for easy parsing. - -## Patterns & Examples - -### View Routing Table -```bash -# Display the current routing table -x route -``` - -### Export to CSV -```bash -# Output the routing table as a CSV -x route --csv -``` - -## Checklist -- [ ] Confirm if the user needs a general view or a structured export. -- [ ] Verify the system environment as routing table formats may vary. diff --git a/data/x-cmd/route/SKILL.zh.md b/data/x-cmd/route/SKILL.zh.md deleted file mode 100644 index 3f071c5..0000000 --- a/data/x-cmd/route/SKILL.zh.md +++ /dev/null @@ -1,37 +0,0 @@ ---- -name: route -description: > - 显示系统的网络路由表,支持结构化数据导出。 - 核心场景:当用户需要检查路由表以进行网络路径诊断时。 -license: MIT ---- - -# route - 网络路由表查看器 - -`route` 模块提供了一个查看系统网络路由表的界面,允许用户检查网络流量的导向。它支持以结构化格式导出路由表。 - -## 激活时机 -- 当诊断网络连通性或路由问题时。 -- 当需要当前路由表的结构化列表 (CSV) 进行分析时。 - -## 核心原则与规范 -- **实验性状态**: 请注意,此模块目前处于实验阶段。 -- **数据导出**: 支持 `--csv` 以便于解析。 - -## 实战示例 - -### 查看路由表 -```bash -# 显示当前路由表 -x route -``` - -### 导出为 CSV -```bash -# 将路由表输出为 CSV -x route --csv -``` - -## 交付验证清单 -- [ ] 确认用户需要的是常规视图还是结构化导出。 -- [ ] 验证系统环境,因为路由表格式可能有所不同。 diff --git a/data/x-cmd/sb/SKILL.md b/data/x-cmd/sb/SKILL.md deleted file mode 100644 index ca59c5d..0000000 --- a/data/x-cmd/sb/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: sb -description: > - SBOM (Software Bill of Materials) utility for generating and analyzing software component lists. - Core Scenario: When the user needs to audit project dependencies or generate SBOM reports for security compliance. -license: MIT ---- - -# sb - SBOM Generator & Analyzer - -The `sb` module provides tools for working with Software Bill of Materials (SBOM), enabling users to generate detailed lists of project components and perform security-focused analysis. - -## When to Activate -- When generating SBOM reports for a project (e.g., using `syft` or `trivy` backends). -- When auditing software dependencies for security compliance. -- When converting between different SBOM formats. - -## Core Principles & Rules -- **Automation**: Designed to integrate into CI/CD pipelines for automated SBOM generation. -- **Backend Support**: Leverages popular SBOM tools under the hood. - -## Patterns & Examples - -### Generate SBOM -```bash -# Create an SBOM report for the current project -x sb gen -``` - -### View Component List -```bash -# List all software components in a human-readable format -x sb ls -``` - -## Checklist -- [ ] Confirm the target project directory. -- [ ] Verify if a specific SBOM format (CycloneDX, SPDX) is required. diff --git a/data/x-cmd/sb/SKILL.zh.md b/data/x-cmd/sb/SKILL.zh.md deleted file mode 100644 index 47bbe78..0000000 --- a/data/x-cmd/sb/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: sb -description: > - SBOM (软件物料清单) 实用工具,用于生成和分析软件组件列表。 - 核心场景:当用户需要审计项目依赖或为安全合规生成 SBOM 报告时。 -license: MIT ---- - -# sb - SBOM 生成与分析器 - -`sb` 模块提供了处理软件物料清单 (SBOM) 的工具,使用户能够生成项目组件的详细列表并执行以安全为中心的分析。 - -## 激活时机 -- 当为项目生成 SBOM 报告时(例如使用 `syft` 或 `trivy` 后端)。 -- 当为了安全合规审计软件依赖项时。 -- 当在不同的 SBOM 格式之间进行转换时。 - -## 核心原则与规范 -- **自动化**: 旨在集成到 CI/CD 流水线中,以实现自动化的 SBOM 生成。 -- **后端支持**: 在底层利用流行的 SBOM 工具。 - -## 实战示例 - -### 生成 SBOM -```bash -# 为当前项目创建一个 SBOM 报告 -x sb gen -``` - -### 查看组件列表 -```bash -# 以人类可读的格式列出所有软件组件 -x sb ls -``` - -## 交付验证清单 -- [ ] 确认目标项目目录。 -- [ ] 验证是否需要特定的 SBOM 格式(CycloneDX, SPDX)。 diff --git a/data/x-cmd/scoop/SKILL.md b/data/x-cmd/scoop/SKILL.md deleted file mode 100644 index dfd4961..0000000 --- a/data/x-cmd/scoop/SKILL.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: scoop -description: > - Enhanced Scoop package manager for Windows, supporting multi-threaded downloads and mirror management. - Core Scenario: When the user needs to manage Windows CLI applications with accelerated downloads via aria2. -license: MIT ---- - -# scoop - Enhanced Windows Package Management - -The `scoop` module provides a powerful interface for the Scoop package manager on Windows, adding support for multi-threaded downloads (via aria2), bucket management, and mirror configuration. - -## When to Activate -- When installing or managing Windows CLI applications. -- When needing to accelerate downloads using multi-threading (`aria2`). -- When managing Scoop buckets or searching for apps across multiple sources. -- When configuring proxies or mirrors for Scoop on Windows. - -## Core Principles & Rules -- **Acceleration**: Encourage using `aria2 enable` for faster package acquisition. -- **Convenience**: Provides an interactive browser (`la`) for discoverability. -- **Clean Environment**: Scoop installs apps to `$HOME/scoop` by default to avoid system clutter. - -## Patterns & Examples - -### Install with Acceleration -```bash -# Enable aria2 and install an app -x scoop aria2 enable -x scoop install telegram -``` - -### Search and List -```bash -# Interactively search for available Scoop packages -x scoop la -``` - -### Bucket Management -```bash -# List all currently added buckets -x scoop bucket list -``` - -## Checklist -- [ ] Confirm if the user is on a Windows environment. -- [ ] Verify if download acceleration (aria2) is desired. -- [ ] Ensure the correct bucket is added for the target application. diff --git a/data/x-cmd/scoop/SKILL.zh.md b/data/x-cmd/scoop/SKILL.zh.md deleted file mode 100644 index 43afbbe..0000000 --- a/data/x-cmd/scoop/SKILL.zh.md +++ /dev/null @@ -1,48 +0,0 @@ ---- -name: scoop -description: > - 适用于 Windows 的增强型 Scoop 包管理器,支持多线程下载和镜像管理。 - 核心场景:当用户需要通过 aria2 加速下载来管理 Windows CLI 应用程序时。 -license: MIT ---- - -# scoop - 增强型 Windows 包管理 - -`scoop` 模块为 Windows 上的 Scoop 包管理器提供了强大的界面,增加了对多线程下载(通过 aria2)、存储桶 (bucket) 管理和镜像配置的支持。 - -## 激活时机 -- 当安装或管理 Windows CLI 应用程序时。 -- 当需要使用多线程 (`aria2`) 加速下载时。 -- 当管理 Scoop 存储桶或在多个来源中搜索应用时。 -- 当在 Windows 上为 Scoop 配置代理或镜像时。 - -## 核心原则与规范 -- **加速**: 鼓励使用 `aria2 enable` 以获得更快的包获取速度。 -- **便捷性**: 提供交互式浏览器 (`la`) 以便进行发现。 -- **清洁环境**: Scoop 默认将应用安装到 `$HOME/scoop`,以避免系统混乱。 - -## 实战示例 - -### 带加速安装 -```bash -# 启用 aria2 并安装一个应用 -x scoop aria2 enable -x scoop install telegram -``` - -### 搜索与罗列 -```bash -# 交互式搜索可用的 Scoop 软件包 -x scoop la -``` - -### 存储桶管理 -```bash -# 列出当前所有已添加的存储桶 -x scoop bucket list -``` - -## 交付验证清单 -- [ ] 确认用户是否处于 Windows 环境。 -- [ ] 验证是否需要下载加速 (aria2)。 -- [ ] 确保已为目标应用程序添加了正确的存储桶。 diff --git a/data/x-cmd/scorecard/SKILL.md b/data/x-cmd/scorecard/SKILL.md deleted file mode 100644 index 42bdf1c..0000000 --- a/data/x-cmd/scorecard/SKILL.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: scorecard -description: > - Automated security tool for assessing open-source project risks and best practices adherence. - Core Scenario: When the user needs to evaluate the security health of a GitHub repository or package. -license: MIT ---- - -# scorecard - OpenSSF Security Scorecard - -The `scorecard` module evaluates open-source projects based on security best practices, providing a score and detailed report on potential risks like binary artifacts, unreviewed code, or dangerous workflows. - -## When to Activate -- When the user wants to assess the security level of an open-source repository. -- When performing due diligence on a new dependency (npm, PyPI, etc.). -- When auditing a local repository for security improvements. - -## Core Principles & Rules -- **Best Practices**: Focuses on identifying risks like lack of CI tests, missing branch protection, or pinned dependencies. -- **Detailed Reporting**: Use `--show-details` to understand why specific checks passed or failed. - -## Patterns & Examples - -### Repository Audit -```bash -# Display the security scorecard for a GitHub repository -x scorecard info github.com/ossf/scorecard -``` - -### Open Web Report -```bash -# Open the full OpenSSF scorecard report in a browser -x scorecard open github.com/owner/repo -``` - -## Checklist -- [ ] Confirm the target repository URL or package name. -- [ ] Verify if the user needs a summary or a detailed check breakdown. diff --git a/data/x-cmd/scorecard/SKILL.zh.md b/data/x-cmd/scorecard/SKILL.zh.md deleted file mode 100644 index a145c34..0000000 --- a/data/x-cmd/scorecard/SKILL.zh.md +++ /dev/null @@ -1,38 +0,0 @@ ---- -name: scorecard -description: > - 用于评估开源项目风险和最佳实践遵循情况的自动化安全工具。 - 核心场景:当用户需要评估 GitHub 仓库或软件包的安全健康状况时。 -license: MIT ---- - -# scorecard - OpenSSF 安全记分卡 - -`scorecard` 模块根据安全最佳实践评估开源项目,提供评分以及关于潜在风险(如二进制工件、未审计的代码或危险的工作流)的详细报告。 - -## 激活时机 -- 当用户想要评估开源仓库的安全级别时。 -- 当对新的依赖项(npm, PyPI 等)执行尽职调查时。 -- 当审计本地仓库以寻求安全改进时。 - -## 核心原则与规范 -- **最佳实践**: 专注于识别缺乏 CI 测试、缺少分支保护或未锁定依赖项等风险。 -- **详细报告**: 使用 `--show-details` 来了解特定检查通过或失败的原因。 - -## 实战示例 - -### 仓库审计 -```bash -# 显示 GitHub 仓库的 OpenSSF 安全记分卡 -x scorecard info github.com/ossf/scorecard -``` - -### 打开网页报告 -```bash -# 在浏览器中打开完整的 OpenSSF 记分卡报告 -x scorecard open github.com/owner/repo -``` - -## 交付验证清单 -- [ ] 确认目标仓库 URL 或软件包名称。 -- [ ] 验证用户需要的是摘要还是详细的检查分解。 diff --git a/data/x-cmd/sd/SKILL.md b/data/x-cmd/sd/SKILL.md deleted file mode 100644 index 0fa9a87..0000000 --- a/data/x-cmd/sd/SKILL.md +++ /dev/null @@ -1,40 +0,0 @@ ---- -name: sd -description: > - Modern finding and replacing tool with intuitive regex syntax and interactive FZF preview. - Core Scenario: When the user needs a faster, more readable alternative to sed for text substitutions. -license: MIT ---- - -# sd - Intuitive Find & Replace Utility - -The `sd` module is a modern alternative to `sed`, focusing on substitution with a more readable syntax. It integrates with FZF for real-time previews of text changes and works without local installation via x-cmd pkg. - -## When to Activate -- When the user wants a more intuitive syntax for text replacement. -- When needing interactive previews of search and replace effects. -- When performing fast string-based replacements without complex sed escaping. - -## Core Principles & Rules -- **Readability**: Prioritize `sd` over `sed` for tasks that only require straightforward substitution. -- **Safety**: Supports `--preview` to output results to stdout without touching the file. -- **Interactive App**: Use `--fzfapp` for visual confirmation of changes. - -## Patterns & Examples - -### Interactive Search and Replace -```bash -# Open interactive FZF preview for text substitution -x sd --fzfapp test.txt -``` - -### String Replacement (Literal) -```bash -# Replace strings literally without treating them as regex -x sd -F "FIND_THIS" "REPLACE_WITH" file.txt -``` - -## Checklist -- [ ] Confirm if the find/replace terms should be treated as literal strings (`-F`). -- [ ] Verify if an interactive preview is required. -- [ ] Ensure the user is aware the file will be modified unless in preview mode. diff --git a/data/x-cmd/sd/SKILL.zh.md b/data/x-cmd/sd/SKILL.zh.md deleted file mode 100644 index 08207ef..0000000 --- a/data/x-cmd/sd/SKILL.zh.md +++ /dev/null @@ -1,40 +0,0 @@ ---- -name: sd -description: > - 现代化的查找和替换工具,具有直观的正则语法和交互式 FZF 预览。 - 核心场景:当用户需要比 sed 更快速、更易读的文本替换方案时。 -license: MIT ---- - -# sd - 直观的查找与替换工具 - -`sd` 模块是 `sed` 的现代化替代方案,专注于使用更易读的语法进行替换。它集成了 FZF 以便实时预览文本更改,并通过 x-cmd pkg 实现免安装运行。 - -## 激活时机 -- 当用户想要更直观的文本替换语法时。 -- 当需要查找和替换效果的交互式预览时。 -- 当执行快速的基于字符串的替换而无需复杂的 sed 转义时。 - -## 核心原则与规范 -- **易读性**: 对于仅需要直接替换的任务,优先选择 `sd` 而非 `sed`。 -- **安全性**: 支持 `--preview`,在不触动文件的情况下将结果输出到标准输出。 -- **交互式应用**: 使用 `--fzfapp` 进行更改的视觉确认。 - -## 实战示例 - -### 交互式查找与替换 -```bash -# 为文本替换打开交互式 FZF 预览 -x sd --fzfapp test.txt -``` - -### 字符串替换 (字面量) -```bash -# 将字符串视为字面量进行替换,而不是正则表达式 -x sd -F "FIND_THIS" "REPLACE_WITH" file.txt -``` - -## 交付验证清单 -- [ ] 确认查找/替换项是否应被视为字面量字符串 (`-F`)。 -- [ ] 验证是否需要交互式预览。 -- [ ] 确保用户知晓除非在预览模式下,否则文件将被修改。 diff --git a/data/x-cmd/se/SKILL.md b/data/x-cmd/se/SKILL.md deleted file mode 100644 index d1f53e6..0000000 --- a/data/x-cmd/se/SKILL.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: se -description: > - Search and browse Stack Exchange (Stack Overflow, Ask Ubuntu, etc.) from the command line. - Core Scenario: When the user needs to find answers to technical questions or browse Stack Overflow solutions. -license: MIT ---- - -# se - Stack Exchange CLI Browser - -The `se` module allows users to search the entire Stack Exchange network, including Stack Overflow and Ask Ubuntu, and browse questions and answers directly in the terminal. - -## When to Activate -- When the user has a specific technical problem and wants to search Stack Overflow. -- When browsing answers for a specific question ID. -- When searching platform-specific sites like Ask Ubuntu for system-related issues. - -## Core Principles & Rules -- **Site Selection**: Support for site-specific prefixes like `:so` (Stack Overflow) and `:au` (Ask Ubuntu). -- **Interactive TUI**: Use `--app` to browse question answers in a structured terminal UI. -- **Search Integration**: Uses DuckDuckGo for broader search within the SE ecosystem if needed. - -## Patterns & Examples - -### Search Stack Overflow -```bash -# Search for JSON parsing issues in Python on Stack Overflow -x se :so "python json parse error" -``` - -### View Answers -```bash -# Get all answers for a specific question ID -x se question --showall 75261408 -``` - -### Interactive App -```bash -# Browse question answers in an interactive TUI -x se question --app 75261408 -``` - -## Checklist -- [ ] Confirm if the search should be limited to a specific SE site (e.g., Stack Overflow). -- [ ] Verify if the user has a specific question ID or needs to perform a keyword search. -- [ ] Check if the interactive view or raw output is preferred. diff --git a/data/x-cmd/se/SKILL.zh.md b/data/x-cmd/se/SKILL.zh.md deleted file mode 100644 index f3166e8..0000000 --- a/data/x-cmd/se/SKILL.zh.md +++ /dev/null @@ -1,46 +0,0 @@ ---- -name: se -description: > - 从命令行搜索和浏览 Stack Exchange(Stack Overflow, Ask Ubuntu 等)。 - 核心场景:当用户需要寻找技术问题的答案或浏览 Stack Overflow 解决方案时。 -license: MIT ---- - -# se - Stack Exchange 命令行浏览器 - -`se` 模块允许用户搜索整个 Stack Exchange 网络(包括 Stack Overflow 和 Ask Ubuntu),并直接在终端中浏览问题和答案。 - -## 激活时机 -- 当用户遇到特定的技术问题并想要搜索 Stack Overflow 时。 -- 当为特定的问题 ID 浏览答案时。 -- 当在 Ask Ubuntu 等特定站点搜索与系统相关的问题时。 - -## 核心原则与规范 -- **站点选择**: 支持特定站点的快捷前缀,如 `:so` (Stack Overflow) 和 `:au` (Ask Ubuntu)。 -- **交互式 TUI**: 使用 `--app` 在结构化终端 UI 中浏览问题的答案。 -- **搜索集成**: 如果需要,使用 DuckDuckGo 在 SE 生态系统中进行更广泛的搜索。 - -## 实战示例 - -### 搜索 Stack Overflow -```bash -# 在 Stack Overflow 上搜索 Python 的 JSON 解析问题 -x se :so "python json parse error" -``` - -### 查看答案 -```bash -# 获取特定问题 ID 的所有答案 -x se question --showall 75261408 -``` - -### 交互式应用 -```bash -# 在交互式 TUI 中浏览问题的答案 -x se question --app 75261408 -``` - -## 交付验证清单 -- [ ] 确认搜索是否应限制在特定的 SE 站点(如 Stack Overflow)。 -- [ ] 验证用户是拥有特定的问题 ID 还是需要执行关键词搜索。 -- [ ] 检查首选的是交互式视图还是原始输出。 diff --git a/data/x-cmd/sed/SKILL.md b/data/x-cmd/sed/SKILL.md deleted file mode 100644 index 434594b..0000000 --- a/data/x-cmd/sed/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: sed -description: > - Enhanced stream editor with interactive preview of search and replace operations using FZF. - Core Scenario: When the user needs to perform text substitutions and wants to see the effects before applying changes. -license: MIT ---- - -# sed - Stream Editor with Interactive Preview - -The `sed` module extends the classic stream editor by adding an interactive preview mode. Users can test their regex substitutions and see context differences (`ctrl-s`) in real-time before modifying files. - -## When to Activate -- When the user wants to perform text search and replace operations. -- When an interactive preview of regex effects is needed to avoid errors. -- When performing bulk line deletions or insertions in files. -- When using extended regular expressions for complex text manipulation. - -## Core Principles & Rules -- **Preview First**: Encourage using the `--fzfapp` mode to verify changes visually. -- **Context Differences**: Use `ctrl-s` within the interactive app to view diffs. -- **In-place Safety**: Use `-i` carefully, preferably with a backup suffix. - -## Patterns & Examples - -### Interactive Preview -```bash -# Interactively test and preview regex changes on a file -x sed --fzfapp test.txt -``` - -### Basic Substitution -```bash -# Replace 'world' with 'universe' in a piped string -echo "hello world" | x sed 's/world/universe/' -``` - -### Global Replace -```bash -# Replace all occurrences of 'foo' with 'bar' in a file -x sed 's/foo/bar/g' file.txt -``` - -## Checklist -- [ ] Confirm if the user needs an interactive preview or direct execution. -- [ ] Verify the regex pattern for search and replace. -- [ ] Check if the file should be modified in-place or output to stdout. diff --git a/data/x-cmd/sed/SKILL.zh.md b/data/x-cmd/sed/SKILL.zh.md deleted file mode 100644 index f1fff57..0000000 --- a/data/x-cmd/sed/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: sed -description: > - 增强型流编辑器,具有使用 FZF 交互式预览查找和替换操作的功能。 - 核心场景:当用户需要执行文本替换并希望在应用更改前查看效果时。 -license: MIT ---- - -# sed - 带有交互式预览的流编辑器 - -`sed` 模块通过添加交互式预览模式扩展了经典的流编辑器。用户可以实时测试其正则表达式替换,并在修改文件前查看上下文差异 (`ctrl-s`)。 - -## 激活时机 -- 当用户想要执行文本查找和替换操作时。 -- 当需要正则表达式效果的交互式预览以避免错误时。 -- 当在文件中执行批量行删除或插入时。 -- 当使用扩展正则表达式进行复杂的文本操作时。 - -## 核心原则与规范 -- **预览先行**: 鼓励使用 `--fzfapp` 模式以直观地验证更改。 -- **上下文差异**: 在交互式应用中使用 `ctrl-s` 查看 diff。 -- **原地安全**: 谨慎使用 `-i`,最好带有备份后缀。 - -## 实战示例 - -### 交互式预览 -```bash -# 在文件上交互式测试并预览正则更改 -x sed --fzfapp test.txt -``` - -### 基础替换 -```bash -# 在管道字符串中将 'world' 替换为 'universe' -echo "hello world" | x sed 's/world/universe/' -``` - -### 全局替换 -```bash -# 全局替换文件中所有出现的 'foo' 为 'bar' -x sed 's/foo/bar/g' file.txt -``` - -## 交付验证清单 -- [ ] 确认用户需要的是交互式预览还是直接执行。 -- [ ] 验证查找和替换的正则模式。 -- [ ] 检查文件应该原地修改还是输出到标准输出。 diff --git a/data/x-cmd/shodan/SKILL.md b/data/x-cmd/shodan/SKILL.md deleted file mode 100644 index 87036fc..0000000 --- a/data/x-cmd/shodan/SKILL.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: shodan -description: > - Search the Shodan network database for connected devices, vulnerabilities, and real-time network intelligence. - Core Scenario: When the user needs to perform network reconnaissance, search for exposed assets, or audit IPs. -license: MIT ---- - -# shodan - Shodan Network Search CLI - -The `shodan` module provides a comprehensive CLI for the Shodan search engine, enabling network reconnaissance, vulnerability detection, and monitoring of internet-connected assets. - -## When to Activate -- When performing reconnaissance on specific IPs or domain names. -- When searching for connected devices with specific vulnerabilities (CVEs) or open ports. -- When monitoring network alerts for specific assets. -- When generating real-time network intelligence summaries via AI integration (`::`). - -## Core Principles & Rules -- **API Key Management**: Remind users to obtain and initialize their Shodan API key via `init`. -- **Targeted Search**: Support for powerful filtering based on facets (ports, protocols, countries). -- **Data Export**: Can download host information and export to structured formats like CSV. - -## Patterns & Examples - -### Scan IPs -```bash -# Perform a targeted scan on specific IP addresses and ports -x shodan scan create 8.8.8.8 1.1.1.1=53/dns-udp,443/https -``` - -### Host Information -```bash -# Lookup IP information in the Shodan internet database -x shodan internetdb 8.8.8.8 -``` - -### AI Intelligence -```bash -# Use AI to summarize Shodan results for specific queries -x shodan :: "critical vulnerabilities in industrial control systems" -``` - -## Checklist -- [ ] Ensure the Shodan API key is configured. -- [ ] Verify the search query or target IP range. -- [ ] Confirm if the user needs raw data or a summarized report. diff --git a/data/x-cmd/shodan/SKILL.zh.md b/data/x-cmd/shodan/SKILL.zh.md deleted file mode 100644 index e0d8662..0000000 --- a/data/x-cmd/shodan/SKILL.zh.md +++ /dev/null @@ -1,47 +0,0 @@ ---- -name: shodan -description: > - 在 Shodan 网络数据库中搜索已连接的设备、漏洞和实时网络情报。 - 核心场景:当用户需要执行网络侦察、搜索暴露的资产或审计 IP 时。 -license: MIT ---- - -# shodan - Shodan 网络搜索命令行 - -`shodan` 模块为 Shodan 搜索引擎提供了全面的命令行界面,支持网络侦察、漏洞检测以及对互联网连接资产的监控。 - -## 激活时机 -- 当对特定的 IP 或域名执行侦察时。 -- 当搜索具有特定漏洞 (CVE) 或开放端口的连接设备时。 -- 当监控特定资产的网络警报时。 -- 当通过 AI 集成 (`::`) 生成实时网络情报摘要时。 - -## 核心原则与规范 -- **API 密钥管理**: 提醒用户通过 `init` 获取并初始化其 Shodan API 密钥。 -- **针对性搜索**: 支持基于维度(端口、协议、国家/地区)的强大过滤功能。 -- **数据导出**: 可以下载主机信息并将其导出为 CSV 等结构化格式。 - -## 实战示例 - -### 扫描 IP -```bash -# 对特定的 IP 地址和端口执行针对性扫描 -x shodan scan create 8.8.8.8 1.1.1.1=53/dns-udp,443/https -``` - -### 主机信息 -```bash -# 在 Shodan 网络数据库中查找 IP 信息 -x shodan internetdb 8.8.8.8 -``` - -### AI 情报 -```bash -# 使用 AI 总结针对特定查询的 Shodan 结果 -x shodan :: "工业控制系统中的关键漏洞" -``` - -## 交付验证清单 -- [ ] 确保已配置 Shodan API 密钥。 -- [ ] 验证搜索查询或目标 IP 范围。 -- [ ] 确认用户需要的是原始数据还是总结报告。 diff --git a/data/x-cmd/sleep/SKILL.md b/data/x-cmd/sleep/SKILL.md deleted file mode 100644 index cd0e761..0000000 --- a/data/x-cmd/sleep/SKILL.md +++ /dev/null @@ -1,52 +0,0 @@ ---- -name: sleep -description: > - Pause execution for human-readable durations or schedule periodic tasks. - Core Scenario: When the user needs to delay scripts, create timers (beacons), or schedule repeating commands. -license: MIT ---- - -# sleep - Pausing & Periodic Scheduling - -The `sleep` module enhances the standard sleep command by supporting human-readable time formats and offering integrated scheduling for repeating tasks. - -## When to Activate -- When a script needs to pause for a duration like `5m` or `1h30m`. -- When creating a periodic "beacon" that outputs timestamps. -- When scheduling a command to run at regular intervals (polling, backups, tests). - -## Core Principles & Rules -- **Time Units**: Use `s`, `m`, `h`, `d` for intuitive duration specification. -- **Combined Command**: Pause then immediately execute a command using `x sleep