From 805afb246092b015421984c80256816e42bc26b3 Mon Sep 17 00:00:00 2001 From: thecodingshrimp Date: Wed, 30 Sep 2026 12:03:43 +0200 Subject: [PATCH 1/6] fix skill dispatch ref: pr-review is a slash command not a skill, replaced with review skill in fix.md --- skills/process/pr-workflow/references/fix.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/skills/process/pr-workflow/references/fix.md b/skills/process/pr-workflow/references/fix.md index 98164f34c..4339b7742 100644 --- a/skills/process/pr-workflow/references/fix.md +++ b/skills/process/pr-workflow/references/fix.md @@ -206,7 +206,7 @@ Solution: ### Related Skills - `pr-pipeline` — For creating PRs from scratch -- `pr-review` — For reviewing code without fixing +- `review` — For reviewing code without fixing (skill); use `/pr-review` command for full multi-agent PR review - `systematic-debugging` — For general debugging unrelated to PR comments ### PR Comment Best Practices From 660bd9ce0dc35f37ed1d4c42fa609259fc59cbb5 Mon Sep 17 00:00:00 2001 From: thecodingshrimp Date: Wed, 30 Sep 2026 13:14:48 +0200 Subject: [PATCH 2/6] recurse into hybrid skill dirs that have both SKILL.md and nested skills so process/* skills appear in index --- scripts/generate-skill-index.py | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/scripts/generate-skill-index.py b/scripts/generate-skill-index.py index 1bc7933c9..b5b5664e3 100755 --- a/scripts/generate-skill-index.py +++ b/scripts/generate-skill-index.py @@ -443,6 +443,17 @@ def _process_skill_dir(skill_dir: Path, skill_file_override: Path | None = None) # Check if this directory directly contains a SKILL.md (flat layout) if (child / "SKILL.md").exists(): _process_skill_dir(child) + # Hybrid directory: also recurse into subdirectories that are skills + # (e.g. skills/process/ has its own SKILL.md AND nested pr-workflow/, workflow/, etc.) + for nested in sorted(child.iterdir()): + if not nested.is_dir(): + continue + if nested.is_symlink() and not include_private: + continue + if (nested / "SKILL.md").exists(): + _process_skill_dir(nested) + elif (nested / "skill" / "SKILL.md").exists(): + _process_skill_dir(nested, skill_file_override=nested / "skill" / "SKILL.md") # Check for nested skill/SKILL.md layout (e.g., voice-example/skill/SKILL.md) elif (child / "skill" / "SKILL.md").exists(): _process_skill_dir(child, skill_file_override=child / "skill" / "SKILL.md") From 4ffcb38b38cea84d059a2c64bf2a6d97ad405a15 Mon Sep 17 00:00:00 2001 From: thecodingshrimp Date: Wed, 30 Sep 2026 14:26:05 +0200 Subject: [PATCH 3/6] fix ref to nonexistent systematic-debugging skill, add hybrid blind spot fix to validate-skill-names, add silent overwrite warning and third-level nesting to generate-skill-index, add hybrid-dir test --- scripts/generate-skill-index.py | 16 ++++++++++-- scripts/tests/test_generate_skill_index.py | 25 +++++++++++++++++++ scripts/validate-skill-names.py | 4 +++ .../toolkit/references/toolkit-evolution.md | 2 +- skills/process/pr-workflow/references/fix.md | 2 +- skills/process/pr-workflow/references/sync.md | 2 +- 6 files changed, 46 insertions(+), 5 deletions(-) diff --git a/scripts/generate-skill-index.py b/scripts/generate-skill-index.py index b5b5664e3..594d74497 100755 --- a/scripts/generate-skill-index.py +++ b/scripts/generate-skill-index.py @@ -430,6 +430,8 @@ def _process_skill_dir(skill_dir: Path, skill_file_override: Path | None = None) old_ok = (repo_root / existing["file"]).is_file() if repo_root else False if old_ok and not new_ok: return + if old_ok and new_ok: + warnings.append(f" - {name}: duplicate skill name — {existing['file']} overwritten by {entry['file']}") index[collection_key][name] = entry for child in sorted(source_dir.iterdir()): @@ -443,8 +445,7 @@ def _process_skill_dir(skill_dir: Path, skill_file_override: Path | None = None) # Check if this directory directly contains a SKILL.md (flat layout) if (child / "SKILL.md").exists(): _process_skill_dir(child) - # Hybrid directory: also recurse into subdirectories that are skills - # (e.g. skills/process/ has its own SKILL.md AND nested pr-workflow/, workflow/, etc.) + # Hybrid dirs (e.g. skills/process/, skills/research/) have both a top-level SKILL.md and nested skill subdirs. for nested in sorted(child.iterdir()): if not nested.is_dir(): continue @@ -454,6 +455,17 @@ def _process_skill_dir(skill_dir: Path, skill_file_override: Path | None = None) _process_skill_dir(nested) elif (nested / "skill" / "SKILL.md").exists(): _process_skill_dir(nested, skill_file_override=nested / "skill" / "SKILL.md") + else: + # Third level within hybrid dir: e.g., skills/process/subcat/{name}/SKILL.md + for deep in sorted(nested.iterdir()): + if not deep.is_dir(): + continue + if deep.is_symlink() and not include_private: + continue + if (deep / "SKILL.md").exists(): + _process_skill_dir(deep) + elif (deep / "skill" / "SKILL.md").exists(): + _process_skill_dir(deep, skill_file_override=deep / "skill" / "SKILL.md") # Check for nested skill/SKILL.md layout (e.g., voice-example/skill/SKILL.md) elif (child / "skill" / "SKILL.md").exists(): _process_skill_dir(child, skill_file_override=child / "skill" / "SKILL.md") diff --git a/scripts/tests/test_generate_skill_index.py b/scripts/tests/test_generate_skill_index.py index 7b115e614..224bc40a9 100755 --- a/scripts/tests/test_generate_skill_index.py +++ b/scripts/tests/test_generate_skill_index.py @@ -376,3 +376,28 @@ def test_include_private_run_writes_only_the_requested_output(self, tmp_path: Pa data = json.loads(output.read_text(encoding="utf-8")) # Floor tracks the post-consolidation catalog; raise it with new skills. assert 35 <= len(data["skills"]) <= 80 + + +class TestHybridDirIndexing: + """Hybrid dirs (top-level SKILL.md + nested skill subdirs) appear fully in index.""" + + def test_hybrid_dir_top_and_nested_both_indexed(self, tmp_path: Path) -> None: + """A hybrid dir's top-level skill and its nested skill both appear in the index.""" + skills_dir = tmp_path / "skills" + # Top-level hybrid skill (e.g. skills/process/SKILL.md) + process_dir = skills_dir / "process" + process_dir.mkdir(parents=True) + (process_dir / "SKILL.md").write_text(_SKILL_FRONTMATTER.format(name="process")) + # Nested skill under the hybrid dir (e.g. skills/process/pr-workflow/SKILL.md) + nested_dir = process_dir / "pr-workflow" + nested_dir.mkdir() + (nested_dir / "SKILL.md").write_text(_SKILL_FRONTMATTER.format(name="pr-workflow")) + + index, _warnings = gsi.generate_index( + source_dir=skills_dir, + dir_prefix="skills", + collection_key="skills", + ) + + assert "process" in index["skills"], "Top-level hybrid skill must be indexed" + assert "pr-workflow" in index["skills"], "Nested skill in hybrid dir must be indexed" diff --git a/scripts/validate-skill-names.py b/scripts/validate-skill-names.py index 5d8db5c2d..9e81f81b4 100644 --- a/scripts/validate-skill-names.py +++ b/scripts/validate-skill-names.py @@ -38,6 +38,10 @@ def find_duplicates(repo: Path) -> dict[str, list[str]]: continue if (top / "SKILL.md").is_file(): seen.setdefault(top.name, []).append(top.relative_to(repo).as_posix()) + # Hybrid dir: also scan immediate subdirs for nested skills. + for child in sorted(top.iterdir()): + if child.is_dir() and (child / "SKILL.md").is_file(): + seen.setdefault(child.name, []).append(child.relative_to(repo).as_posix()) continue for child in sorted(top.iterdir()): if child.is_dir() and (child / "SKILL.md").is_file(): diff --git a/skills/meta/toolkit/references/toolkit-evolution.md b/skills/meta/toolkit/references/toolkit-evolution.md index dfd708b76..c48caac88 100644 --- a/skills/meta/toolkit/references/toolkit-evolution.md +++ b/skills/meta/toolkit/references/toolkit-evolution.md @@ -238,7 +238,7 @@ Win condition for each implementation: **Step 1: Handle winners (WIN status)** -For each winning implementation, create a PR using the template from `references/evolve-scripts.md` § Step 1, then merge. After creating the PR, run pr-review to validate, then merge. +For each winning implementation, create a PR using the template from `references/evolve-scripts.md` § Step 1, then merge. After creating the PR, run /pr-review to validate, then merge. The multi-persona critique + A/B testing gate is the review. Auto-merge is safe because the validation happened before this step. diff --git a/skills/process/pr-workflow/references/fix.md b/skills/process/pr-workflow/references/fix.md index 4339b7742..7c5249557 100644 --- a/skills/process/pr-workflow/references/fix.md +++ b/skills/process/pr-workflow/references/fix.md @@ -207,7 +207,7 @@ Solution: ### Related Skills - `pr-pipeline` — For creating PRs from scratch - `review` — For reviewing code without fixing (skill); use `/pr-review` command for full multi-agent PR review -- `systematic-debugging` — For general debugging unrelated to PR comments +- `debugging` — For general debugging unrelated to PR comments ### PR Comment Best Practices - Always validate before fixing (this prevents introducing bugs that reviewers caught) diff --git a/skills/process/pr-workflow/references/sync.md b/skills/process/pr-workflow/references/sync.md index 95bb8ff26..2af2f3c64 100644 --- a/skills/process/pr-workflow/references/sync.md +++ b/skills/process/pr-workflow/references/sync.md @@ -221,7 +221,7 @@ Solution: ## References -- `/pr-review` -- Comprehensive PR review (used in the review-fix loop) +- `/pr-review` -- Comprehensive PR review slash command (not the `review` skill; used in the review-fix loop) - `/pr-cleanup` -- Post-merge branch cleanup - `scripts/classify-repo.py` -- Repo classification for workflow gating - `scripts/adr-decision-coverage.py` -- ADR decision coverage checker From b50ed056f9b2c7522b006d4e9f7d6aadcb648617 Mon Sep 17 00:00:00 2001 From: thecodingshrimp Date: Wed, 30 Sep 2026 17:28:17 +0200 Subject: [PATCH 4/6] regenerate routing-map.md to sync agent count and descriptions with current frontmatter --- docs/routing-map.md | 98 ++++++++++++++++++++++----------------------- 1 file changed, 49 insertions(+), 49 deletions(-) diff --git a/docs/routing-map.md b/docs/routing-map.md index a9973aed2..0f66e959e 100644 --- a/docs/routing-map.md +++ b/docs/routing-map.md @@ -6,58 +6,58 @@ Single source of truth: agent/skill frontmatter and `pipeline-index.json`. Regenerate: `python3 scripts/generate-routing-map.py`. CI check: `python3 scripts/generate-routing-map.py --check`. -## AGENTS (42) +## AGENTS (43) | name | when to call (description) | triggers | FORCE | not_for | pairs_with | |---|---|---|---|---|---| -| ansible-automation-engineer | Ansible automation: playbooks, roles, collections, Molecule testing, Vault security | ansible, playbook, automation, molecule, ansible-tower, AWX | | Kubernetes manifests, Helm charts, or cluster deploys (use kubernetes-helm-engineer); Claude Code hook or toolkit automation (use hook-development-engineer); application test automation (use testing-automation-engineer). This agent writes Ansible playbooks, roles, collections, and Molecule tests. | testing, kubernetes-helm-engineer | -| combat-effects-upgrade | Zero-dependency combat visual upgrades: CSS particle replacement, Framer Motion combat juice, CSS 3D card transforms | combat effects, CSS particles, particle replacement, framer motion combat, combat animations, card transforms, CSS 3D, combat juice, visual effects upgrade, effects.ts, combat polish | | WebGL or GPU-particle combat rendering (use pixijs-combat-renderer); skeletal or character animation rigs (use rive-skeletal-animator); general frontend-system, layout, or accessibility work (use ui-design-engineer). This agent does zero-dependency CSS and Framer Motion combat effects. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | -| data-engineer | Data pipelines, ETL/ELT, warehouse frontend, dimensional modeling, stream processing | data pipeline, ETL, ELT, dbt, Airflow, Prefect, Dagster, dimensional model, data warehouse, star schema, snowflake schema, data lake, data quality, streaming, Kafka, Spark, Flink, BigQuery, Redshift, Parquet, Delta Lake, Iceberg, data vault, slowly changing dimension, SCD, data lineage | | OLTP schema/query tuning (use database-engineer); RabbitMQ or AMQP broker topology, clustering, and high availability (use rabbitmq-messaging-engineer) | database-engineer, data | -| database-engineer | Database frontend, optimization, query performance, migrations, indexing strategies | database, schema, SQL, postgres, mysql, indexing, query optimization | | embedded SQLite or Peewee ORM work (use sqlite-peewee-engineer) — this agent's foreign-key and zero-downtime migration mandates assume a server database and misfire on embedded SQLite; analytics warehouses, ETL/ELT, dimensional models, or stream processing (use data-engineer); OpenSearch or Elasticsearch cluster tuning (use opensearch-elasticsearch-engineer). This agent frontends and tunes server databases such as Postgres and MySQL. | nodejs-api-engineer, sqlite-peewee-engineer, data-engineer | -| github-engineer | Extract coding conventions and style rules from GitHub user profiles via API | github rules, profile analysis, coding style extraction, github conventions, programming rules | | deriving conventions from a local repository (use assessment skill); generating a project CLAUDE.md from repo analysis (use toolkit skill); building a writing-voice profile from author samples (use writing skill); reviewing code against conventions already adopted (use reviewer-code). This agent extracts coding rules from GitHub user profiles via the API. | assessment, toolkit | -| golang-general-engineer | Go development: features, debugging, code review, performance | go, golang, .go files, gofmt, go mod, goroutine, channel, gopls | | tasks using 'go' as a verb (go ahead, go fix this); Kotlin coroutine work (use kotlin-general-engineer); Go concurrency patterns in isolation (use programming skill) | programming | -| hook-development-engineer | Python hook development for the Claude Code event-driven system and its telemetry store | create hook, hook development, event handler, PostToolUse, PreToolUse, SessionStart, routing telemetry hook, hook registration | | auditing hook conventions, registration, or timeouts fleet-wide (use toolkit-governance-engineer); adapting existing hooks to a new Claude Code release (use system-upgrade-engineer); reading routing telemetry that hooks already recorded (use the process skill); general Python features outside the hook system (use python-general-engineer). This agent writes and debugs Python hook implementations. | testing, code-quality | -| kotlin-general-engineer | Kotlin development: features, coroutines, debugging, code quality, multiplatform | programming, ktor, koin, coroutine, suspend fun, programming flow, StateFlow, kotest, mockk, gradle-kts, detekt, ktlint, ktfmt, android programming, programming-multiplatform | | writing Kotlin tests (use programming); coroutine or Flow patterns in isolation (use programming); Go goroutine work (use golang-general-engineer). This agent writes and debugs Kotlin features. | workflow, testing, review | -| kubernetes-helm-engineer | Kubernetes and Helm: deployments, troubleshooting, cloud-native infrastructure | kubernetes, helm, k8s, kubectl, statefulset, argocd, deployment | | Ansible playbooks and host configuration management (use ansible-automation-engineer); Prometheus or Grafana monitoring, alerting, and PromQL (use prometheus-grafana-engineer); Perses dashboards, plugins, or operator work (use perses-engineer); cluster debugging, RBAC, and security operations (use kubernetes skill). This agent authors and fixes Kubernetes manifests and Helm charts. | testing, prometheus-grafana-engineer | -| mcp-local-docs-engineer | MCP server development for local documentation access in TypeScript/Node | MCP, docs server, documentation server, hugo | | using existing MCP tools/servers (only building local-docs MCP servers); web performance or LCP optimization (use performance-optimization-engineer); writing the documentation content itself, API references, or runbooks (use technical-documentation-engineer) | testing | -| nextjs-ecommerce-engineer | Use this agent when building a NextJS e-commerce site: shopping cart, Stripe payments, product catalogs, order management, and checkout flows | next.js e-commerce, nextjs ecommerce, shopping cart, stripe, e-commerce, online store, product catalog | | general React or Next.js architecture with no commerce concern (use typescript-frontend-engineer); portfolio and gallery sites (use react-portfolio-engineer); visual frontend systems and layout (use ui-design-engineer). This agent builds Next.js commerce: carts, Stripe checkout, catalogs, and orders. | testing, typescript-frontend-engineer | -| nodejs-api-engineer | Use this agent when you need expert assistance with NodeJS backend API development: REST endpoints, authentication, file uploads, webhooks, middleware, and database integration | node.js, nodejs, express, API, backend, webhook, authentication | | KPI dashboards or data analysis (use data skill) — this agent builds Node.js REST/GraphQL APIs | review, database-engineer | -| opensearch-elasticsearch-engineer | OpenSearch/Elasticsearch: cluster management, performance tuning, index optimization | opensearch, elasticsearch, search cluster, logstash, kibana, search performance | | search-relevance UX or query-DSL authoring inside an app (use domain skill); relational database indexing and query tuning (use database-engineer). This agent runs and tunes the OpenSearch or Elasticsearch cluster itself. | testing, domain | -| performance-optimization-engineer | Web performance optimization: Core Web Vitals, rendering, bundle analysis, monitoring | core web vitals, LCP, bundle size, render performance, web performance, lighthouse | | backend/algorithmic performance in a specific language (use that language engineer); MCP server development (use mcp-local-docs-engineer) — only web/frontend runtime and load performance | testing | +| ansible-automation-engineer | Ansible automation: playbooks, roles, collections, Molecule testing, Vault security | ansible, playbook, automation, molecule, ansible-tower, AWX | | Kubernetes manifests, Helm charts, or cluster deploys (use kubernetes-helm-engineer); Claude Code hook or toolkit automation (use hook-development-engineer); application test automation (use testing-automation-engineer). This agent writes Ansible playbooks, roles, collections, and Molecule tests. | verification-before-completion, kubernetes-helm-engineer | +| combat-effects-upgrade | Zero-dependency combat visual upgrades: CSS particle replacement, Framer Motion combat juice, CSS 3D card transforms | combat effects, CSS particles, particle replacement, framer motion combat, combat animations, card transforms, CSS 3D, combat juice, visual effects upgrade, effects.ts, combat polish | | WebGL or GPU-particle combat rendering (use pixijs-combat-renderer); skeletal or character animation rigs (use rive-skeletal-animator); general design-system, layout, or accessibility work (use ui-design-engineer). This agent does zero-dependency CSS and Framer Motion combat effects. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | +| data-engineer | Data pipelines, ETL/ELT, warehouse design, dimensional modeling, stream processing | data pipeline, ETL, ELT, dbt, Airflow, Prefect, Dagster, dimensional model, data warehouse, star schema, snowflake schema, data lake, data quality, streaming, Kafka, Spark, Flink, BigQuery, Redshift, Parquet, Delta Lake, Iceberg, data vault, slowly changing dimension, SCD, data lineage | | OLTP schema/query tuning (use database-engineer); RabbitMQ or AMQP broker topology, clustering, and high availability (use rabbitmq-messaging-engineer) | database-engineer, data-analysis | +| database-engineer | Database design, optimization, query performance, migrations, indexing strategies | database, schema, SQL, postgres, mysql, indexing, query optimization | | embedded SQLite or Peewee ORM work (use sqlite-peewee-engineer) — this agent's foreign-key and zero-downtime migration mandates assume a server database and misfire on embedded SQLite; analytics warehouses, ETL/ELT, dimensional models, or stream processing (use data-engineer); OpenSearch or Elasticsearch cluster tuning (use opensearch-elasticsearch-engineer). This agent designs and tunes server databases such as Postgres and MySQL. | nodejs-api-engineer, sqlite-peewee-engineer, data-engineer | +| github-profile-rules-engineer | Extract coding conventions and style rules from GitHub user profiles via API | github rules, profile analysis, coding style extraction, github conventions, programming rules | | deriving conventions from a local repository (use codebase-overview skill); generating a project CLAUDE.md from repo analysis (use generate-claudemd skill); building a writing-voice profile from author samples (use create-voice skill); reviewing code against conventions already adopted (use reviewer-code). This agent extracts coding rules from GitHub user profiles via the API. | codebase-overview, generate-claudemd | +| golang-general-engineer | Go development: features, debugging, code review, performance | go, golang, .go files, gofmt, go mod, goroutine, channel, gopls | | tasks using 'go' as a verb (go ahead, go fix this); Kotlin coroutine work (use kotlin-general-engineer); Go concurrency patterns in isolation (use go-patterns skill) | go-patterns | +| golang-general-engineer-compact | Compact Go development for tight context budgets | tight context, compact, focused go | | general Go feature/debug work via /do (use golang-general-engineer) — only tight-context budgets and per-package review fan-out | go-patterns | +| hook-development-engineer | Python hook development for Claude Code event-driven system and learning database | create hook, hook development, event handler, PostToolUse, PreToolUse, SessionStart, learning database, error detection hook | | auditing hook conventions, registration, or timeouts fleet-wide (use toolkit-governance-engineer); adapting existing hooks to a new Claude Code release (use system-upgrade-engineer); querying or graduating captured learnings (use learn skill); general Python features outside the hook system (use python-general-engineer). This agent writes and debugs Python hook implementations. | verification-before-completion, python-quality-gate | +| kotlin-general-engineer | Kotlin development: features, coroutines, debugging, code quality, multiplatform | kotlin, ktor, koin, coroutine, suspend fun, kotlin flow, StateFlow, kotest, mockk, gradle-kts, detekt, ktlint, ktfmt, android kotlin, kotlin-multiplatform | | writing Kotlin tests (use kotlin); coroutine or Flow patterns in isolation (use kotlin); Go goroutine work (use golang-general-engineer). This agent writes and debugs Kotlin features. | workflow, verification-before-completion, systematic-code-review | +| kubernetes-helm-engineer | Kubernetes and Helm: deployments, troubleshooting, cloud-native infrastructure | kubernetes, helm, k8s, kubectl, statefulset, argocd, deployment | | Ansible playbooks and host configuration management (use ansible-automation-engineer); Prometheus or Grafana monitoring, alerting, and PromQL (use prometheus-grafana-engineer); Perses dashboards, plugins, or operator work (use perses-engineer); cluster debugging, RBAC, and security operations (use kubernetes skill). This agent authors and fixes Kubernetes manifests and Helm charts. | verification-before-completion, prometheus-grafana-engineer | +| mcp-local-docs-engineer | MCP server development for local documentation access in TypeScript/Node | MCP, docs server, documentation server, hugo | | using existing MCP tools/servers (only building local-docs MCP servers); web performance or LCP optimization (use performance-optimization-engineer); writing the documentation content itself, API references, or runbooks (use technical-documentation-engineer) | verification-before-completion | +| nextjs-ecommerce-engineer | Use this agent when building a NextJS e-commerce site: shopping cart, Stripe payments, product catalogs, order management, and checkout flows | next.js e-commerce, nextjs ecommerce, shopping cart, stripe, e-commerce, online store, product catalog | | general React or Next.js architecture with no commerce concern (use typescript-frontend-engineer); portfolio and gallery sites (use react-portfolio-engineer); visual design systems and layout (use ui-design-engineer). This agent builds Next.js commerce: carts, Stripe checkout, catalogs, and orders. | verification-before-completion, typescript-frontend-engineer | +| nodejs-api-engineer | Use this agent when you need expert assistance with NodeJS backend API development: REST endpoints, authentication, file uploads, webhooks, middleware, and database integration | node.js, nodejs, express, API, backend, webhook, authentication | | KPI dashboards or data analysis (use data-analysis skill) — this agent builds Node.js REST/GraphQL APIs | systematic-code-review, database-engineer | +| opensearch-elasticsearch-engineer | OpenSearch/Elasticsearch: cluster management, performance tuning, index optimization | opensearch, elasticsearch, search cluster, logstash, kibana, search performance | | search-relevance UX or query-DSL authoring inside an app (use enterprise-search); relational database indexing and query tuning (use database-engineer). This agent runs and tunes the OpenSearch or Elasticsearch cluster itself. | verification-before-completion, enterprise-search | +| performance-optimization-engineer | Web performance optimization: Core Web Vitals, rendering, bundle analysis, monitoring | core web vitals, LCP, bundle size, render performance, web performance, lighthouse | | backend/algorithmic performance in a specific language (use that language engineer); MCP server development (use mcp-local-docs-engineer) — only web/frontend runtime and load performance | verification-before-completion | | perses-engineer | Perses observability platform: dashboards, plugins, operator, core development | perses, perses dashboard, perses plugin, perses operator, perses kubernetes, perses CRD, percli, perses migrate, perses dac, PersesDashboard, perses core, contribute perses, perses backend, perses frontend, perses architecture, perses internals, perses project, observability dashboard, perses datasource, perses variable, perses helm, perses k8s, create plugin, panel plugin, datasource plugin, perses plugin development, perses cue schema, plugin schema | | Grafana or Prometheus dashboards, alerting, and PromQL (use prometheus-grafana-engineer); Kubernetes manifests and Helm charts unrelated to Perses (use kubernetes-helm-engineer); TypeScript frontend work outside the Perses codebase (use typescript-frontend-engineer). This agent develops the Perses platform itself: dashboards, plugins, operator, and core. | golang-general-engineer, typescript-frontend-engineer, kubernetes-helm-engineer, prometheus-grafana-engineer | -| php-general-engineer | PHP development: features, debugging, code quality, security, modern PHP 8 | programming, laravel, symfony, composer, artisan, eloquent, blade, twig, programmingunit, pest, psr-12, psr standards, hybris, sapcc, .programming files, doctrine, programming-cs-fixer, programmingstan, psalm | | running or configuring PHP quality or test tooling in isolation (use programming skill); Go SAP Commerce Cloud patterns (use programming skill); Python pip/packaging (use python-general-engineer). This agent writes and debugs PHP features. | workflow, testing, review | -| pipeline-orchestrator-engineer | Pipeline orchestration: scaffold multi-component workflows, fan-out/fan-in patterns | create pipeline, new pipeline, scaffold pipeline, build pipeline, pipeline creator | | coordinating agents and dependencies across an in-flight project (use project-coordinator-engineer); authoring one new skill end to end (use toolkit skill); running an existing workflow or DAG (use workflow skill); routing-table and INDEX consistency (use toolkit-governance-engineer). This agent scaffolds new multi-component pipelines with fan-out/fan-in structure. | workflow, assessment, toolkit | -| pixijs-combat-renderer | PixiJS v8 2D WebGL combat rendering: @pixi/react hybrid canvas, normal maps, GPU particles, post-processing | pixijs, pixi.js, pixi react, @pixi/react, 2D WebGL, GPU particles, combat renderer, normal map 2D, pixi filters, sprite rendering GPU | | CSS or Framer Motion effects that need no WebGL (use combat-effects-upgrade); skeletal character animation and state machines (use rive-skeletal-animator); 3D or Three.js scenes (use frontend skill); general React component architecture (use typescript-frontend-engineer). This agent renders 2D WebGL combat with PixiJS v8. | typescript-frontend-engineer, ui-design-engineer, combat-effects-upgrade | -| project-coordinator-engineer | Multi-agent project coordination: task breakdown, dependency management, progress tracking | coordinate, multi-agent, orchestrate, project, task management, agent coordination | | personal productivity and task-list tools (use business-ops skill, PRODUCTIVITY mode); scaffolding the structure of a new multi-component pipeline (use pipeline-orchestrator-engineer) — this agent coordinates multi-agent project execution and dependency tracking | workflow, process | -| prometheus-grafana-engineer | Prometheus and Grafana: monitoring, alerting, dashboard frontend, PromQL optimization | prometheus, grafana, monitoring, alerting, dashboards, metrics, observability | | analyzing business or product metrics data (use data skill); Perses dashboards, plugins, operator, or core development (use perses-engineer); deploying the monitoring stack via Kubernetes manifests or Helm charts (use kubernetes-helm-engineer). This agent is Prometheus and Grafana operations, not data analytics. | testing, kubernetes-helm-engineer | -| python-general-engineer | Python development: features, debugging, code review, performance | python, .py files, pip, pytest, asyncio, fastapi, django, flask | | PHP development (use php-general-engineer); OpenStack services and Oslo libraries (use python-openstack-engineer); SQLite and Peewee ORM work (use sqlite-peewee-engineer) — this agent handles general Python development | code-quality | -| python-openstack-engineer | OpenStack Python development: Nova, Neutron, Cinder, Oslo libraries, WSGI middleware | openstack, oslo, neutron, nova, cinder, tempest, oslo.config, oslo.messaging | | general Python features, debugging, or web frameworks outside OpenStack (use python-general-engineer); running Python lint, format, and test gates (use code-quality skill); SQLite and Peewee ORM work (use sqlite-peewee-engineer). This agent develops OpenStack services and Oslo libraries. | code-quality, python-general-engineer | -| rabbitmq-messaging-engineer | RabbitMQ: message queue architecture, clustering, high-availability, routing patterns | rabbitmq, messaging, message queue, amqp, event bus | | Kafka, Spark, or stream-processing data pipelines (use data-engineer); application webhook and REST endpoint code (use nodejs-api-engineer); running a broker on Kubernetes via manifests or Helm (use kubernetes-helm-engineer); broker metrics dashboards and alert rules (use prometheus-grafana-engineer). This agent frontends RabbitMQ topology, clustering, and high availability. | testing | -| react-native-engineer | React Native and Expo development: performance, animations, navigation, native UI patterns | react native, expo, reanimated, flashlist, hermes, react-navigation, gesture handler, native stack | | React web apps (use typescript-frontend-engineer); portfolio and gallery sites for creatives (use react-portfolio-engineer) | code-quality | +| php-general-engineer | PHP development: features, debugging, code quality, security, modern PHP 8 | php, laravel, symfony, composer, artisan, eloquent, blade, twig, phpunit, pest, psr-12, psr standards, hybris, sapcc, .php files, doctrine, php-cs-fixer, phpstan, psalm | | running or configuring PHP quality or test tooling in isolation (use php skill); Go SAP Commerce Cloud patterns (use go-patterns skill); Python pip/packaging (use python-general-engineer). This agent writes and debugs PHP features. | workflow, verification-before-completion, systematic-code-review | +| pipeline-orchestrator-engineer | Pipeline orchestration: scaffold multi-component workflows, fan-out/fan-in patterns | create pipeline, new pipeline, scaffold pipeline, build pipeline, pipeline creator | | coordinating agents and dependencies across an in-flight project (use project-coordinator-engineer); authoring one new skill end to end (use skill-creator skill); running an existing workflow or DAG (use workflow skill); routing-table and INDEX consistency (use toolkit-governance-engineer). This agent scaffolds new multi-component pipelines with fan-out/fan-in structure. | workflow, codebase-overview, routing-table-updater | +| pixijs-combat-renderer | PixiJS v8 2D WebGL combat rendering: @pixi/react hybrid canvas, normal maps, GPU particles, post-processing | pixijs, pixi.js, pixi react, @pixi/react, 2D WebGL, GPU particles, combat renderer, normal map 2D, pixi filters, sprite rendering GPU | | CSS or Framer Motion effects that need no WebGL (use combat-effects-upgrade); skeletal character animation and state machines (use rive-skeletal-animator); 3D or Three.js scenes (use threejs-builder skill); general React component architecture (use typescript-frontend-engineer). This agent renders 2D WebGL combat with PixiJS v8. | typescript-frontend-engineer, ui-design-engineer, combat-effects-upgrade | +| project-coordinator-engineer | Multi-agent project coordination: task breakdown, dependency management, progress tracking | coordinate, multi-agent, orchestrate, project, task management, agent coordination | | personal productivity and task-list tools (use business-ops skill, PRODUCTIVITY mode); scaffolding the structure of a new multi-component pipeline (use pipeline-orchestrator-engineer) — this agent coordinates multi-agent project execution and dependency tracking | workflow, subagent-driven-development | +| prometheus-grafana-engineer | Prometheus and Grafana: monitoring, alerting, dashboard design, PromQL optimization | prometheus, grafana, monitoring, alerting, dashboards, metrics, observability | | analyzing business or product metrics data (use data-analysis); Perses dashboards, plugins, operator, or core development (use perses-engineer); deploying the monitoring stack via Kubernetes manifests or Helm charts (use kubernetes-helm-engineer). This agent is Prometheus and Grafana operations, not data analytics. | verification-before-completion, kubernetes-helm-engineer | +| python-general-engineer | Python development: features, debugging, code review, performance | python, .py files, pip, pytest, asyncio, fastapi, django, flask | | PHP development (use php-general-engineer); OpenStack services and Oslo libraries (use python-openstack-engineer); SQLite and Peewee ORM work (use sqlite-peewee-engineer) — this agent handles general Python development | python-quality-gate | +| python-openstack-engineer | OpenStack Python development: Nova, Neutron, Cinder, Oslo libraries, WSGI middleware | openstack, oslo, neutron, nova, cinder, tempest, oslo.config, oslo.messaging | | general Python features, debugging, or web frameworks outside OpenStack (use python-general-engineer); running Python lint, format, and test gates (use python-quality-gate skill); SQLite and Peewee ORM work (use sqlite-peewee-engineer). This agent develops OpenStack services and Oslo libraries. | python-quality-gate, python-general-engineer | +| rabbitmq-messaging-engineer | RabbitMQ: message queue architecture, clustering, high-availability, routing patterns | rabbitmq, messaging, message queue, amqp, event bus | | Kafka, Spark, or stream-processing data pipelines (use data-engineer); application webhook and REST endpoint code (use nodejs-api-engineer); running a broker on Kubernetes via manifests or Helm (use kubernetes-helm-engineer); broker metrics dashboards and alert rules (use prometheus-grafana-engineer). This agent designs RabbitMQ topology, clustering, and high availability. | verification-before-completion | +| react-native-engineer | React Native and Expo development: performance, animations, navigation, native UI patterns | react native, expo, reanimated, flashlist, hermes, react-navigation, gesture handler, native stack | | React web apps (use typescript-frontend-engineer); portfolio and gallery sites for creatives (use react-portfolio-engineer) | universal-quality-gate | | react-portfolio-engineer | React portfolio/gallery sites for creatives: React 18+, Next | portfolio, gallery, react portfolio, art website, image gallery, lightbox | | commerce features such as carts, Stripe, or checkout (use nextjs-ecommerce-engineer); general React or Next.js application architecture (use typescript-frontend-engineer); React Native mobile apps (use react-native-engineer); generating the artwork itself (use image-gen skill). This agent builds portfolio and gallery sites for creatives. | ui-design-engineer, typescript-frontend-engineer | -| research-coordinator-engineer | Research coordination: systematic investigation, multi-source analysis, synthesis | research topic, multi-source analysis, investigate question, synthesize sources, comprehensive analysis, study, examine | | reading/exploring local code (use assessment skill); a single delegated research subtask (use research-subagent-executor); running the formal multi-phase research pipeline (use research skill) | workflow, process | -| research-subagent-executor | Research subagent execution: OODA-loop investigation, intelligence gathering, source evaluation | research subtask, delegated research | | decomposing a question or synthesizing across many sources (use research-coordinator-engineer); running the full multi-phase research pipeline (use research skill); exploring local code (use assessment skill). This agent executes one delegated research subtask via OODA loops. | research-coordinator-engineer | -| reviewer-code | Code quality review: conventions, naming, dead code, performance, test coverage | code review, review code quality, code conventions, naming review, dead code review, performance review, type frontend review, test coverage review, config safety review | | Codex second-opinion review or PR lifecycle (use pr-workflow skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent reviews code quality: conventions, naming, dead code, test coverage | workflow, review | -| reviewer-domain | Domain-specific review: ADR compliance, business logic, SAP CC structural, pragmatic builder | adr compliance, adr review, architecture decision, decision record, adr check, scope creep, business logic, domain review, requirements, correctness, edge cases, state machine, sapcc structural, go-bits frontend, sapcc structural review, type export, anti-over-engineering, go-bits usage, builder, production, ops, operational | | naming, dead code, test coverage, and convention review (use reviewer-code); security, concurrency, error handling, observability, and API contracts (use reviewer-system); newcomer, contrarian, pedant, or user-advocate framings (use reviewer-perspectives); full-repo SAP Commerce Cloud compliance audit (use domain skill); running a consultation to decide a new architecture decision (use assessment skill). This agent reviews business logic, ADR conformance, and SAP CC structural correctness against domain requirements. | workflow, review | -| reviewer-perspectives | Multi-perspective review: newcomer, senior, pedant, contrarian, user advocate, meta-process | newcomer perspective, fresh eyes review, documentation review, production readiness, senior review, skeptical review, technical accuracy, spec compliance, pedantic review, contrarian, alternatives, assumptions, challenge, assessment, user impact, user advocate, usability review, is this worth the complexity, user perspective, user experience, meta-process review, system frontend review, architecture health, single point of failure, indispensable component, complexity audit, authority concentration, reversibility check | | SAP Commerce Cloud full-repo compliance audit (use domain skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides multi-perspective review: newcomer, senior, pedant, contrarian views | review, workflow | -| reviewer-system | System-level review: security, concurrency, error handling, observability, API contracts | system review, security review, concurrency review, error handling review, observability review, API contract review, migration safety review, dependency audit, API doc accuracy review | | standalone security scan of git changes (use security skill); parallel multi-reviewer code review (use review skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides system-level review: concurrency, error handling, observability, API contracts | workflow, review, programming | -| rive-skeletal-animator | Rive skeletal animation: @rive-app/react-canvas, state machines, character pipelines, combat integration | rive, rive-app, skeletal animation, character animation, spine animation, animation state machine, idle animation, hit reaction animation, @rive-app/react-canvas, .riv files | | GPU particle and WebGL combat rendering (use pixijs-combat-renderer); CSS or Framer Motion effects (use combat-effects-upgrade); 2D sprite-sheet generation pipelines (use game-dev skill); 3D scene animation (use frontend skill). This agent integrates Rive skeletal animation, state machines, and character pipelines. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | +| research-coordinator-engineer | Research coordination: systematic investigation, multi-source analysis, synthesis | research topic, multi-source analysis, investigate question, synthesize sources, comprehensive analysis, study, examine | | reading/exploring local code (use codebase-overview/read-only-ops); a single delegated research subtask (use research-subagent-executor); running the formal multi-phase research pipeline (use research-pipeline skill) | workflow, subagent-driven-development | +| research-subagent-executor | Research subagent execution: OODA-loop investigation, intelligence gathering, source evaluation | research subtask, delegated research | | decomposing a question or synthesizing across many sources (use research-coordinator-engineer); running the full multi-phase research pipeline (use research-pipeline skill); exploring local code (use codebase-overview skill). This agent executes one delegated research subtask via OODA loops. | research-coordinator-engineer | +| reviewer-code | Code quality review: conventions, naming, dead code, performance, test coverage | code review, review code quality, code conventions, naming review, dead code review, performance review, type design review, test coverage review, config safety review | | Codex second-opinion review or PR lifecycle (use pr-workflow skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent reviews code quality: conventions, naming, dead code, test coverage | workflow, parallel-code-review, systematic-code-review | +| reviewer-domain | Domain-specific review: ADR compliance, business logic, SAP CC structural, pragmatic builder | adr compliance, adr review, architecture decision, decision record, adr check, scope creep, business logic, domain review, requirements, correctness, edge cases, state machine, sapcc structural, go-bits design, sapcc structural review, type export, anti-over-engineering, go-bits usage, builder, production, ops, operational | | naming, dead code, test coverage, and convention review (use reviewer-code); security, concurrency, error handling, observability, and API contracts (use reviewer-system); newcomer, contrarian, pedant, or user-advocate framings (use reviewer-perspectives); full-repo SAP Commerce Cloud compliance audit (use sapcc-audit skill); running a consultation to decide a new architecture decision (use adr-consultation skill). This agent reviews business logic, ADR conformance, and SAP CC structural correctness against domain requirements. | workflow, parallel-code-review, systematic-code-review | +| reviewer-perspectives | Multi-perspective review: newcomer, senior, pedant, contrarian, user advocate, meta-process | newcomer perspective, fresh eyes review, documentation review, production readiness, senior review, skeptical review, technical accuracy, spec compliance, pedantic review, contrarian, alternatives, assumptions, challenge, multi-persona-critique, user impact, user advocate, usability review, is this worth the complexity, user perspective, user experience, meta-process review, system design review, architecture health, single point of failure, indispensable component, complexity audit, authority concentration, reversibility check | | SAP Commerce Cloud full-repo compliance audit (use sapcc-audit skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides multi-perspective review: newcomer, senior, pedant, contrarian views | systematic-code-review, workflow | +| reviewer-system | System-level review: security, concurrency, error handling, observability, API contracts | system review, security review, concurrency review, error handling review, observability review, API contract review, migration safety review, dependency audit, API doc accuracy review | | standalone security scan of git changes (use security-review skill); parallel multi-reviewer code review (use parallel-code-review skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides system-level review: concurrency, error handling, observability, API contracts | workflow, systematic-code-review, parallel-code-review, go-patterns | +| rive-skeletal-animator | Rive skeletal animation: @rive-app/react-canvas, state machines, character pipelines, combat integration | rive, rive-app, skeletal animation, character animation, spine animation, animation state machine, idle animation, hit reaction animation, @rive-app/react-canvas, .riv files | | GPU particle and WebGL combat rendering (use pixijs-combat-renderer); CSS or Framer Motion effects (use combat-effects-upgrade); 2D sprite-sheet generation pipelines (use game-sprite-pipeline skill); 3D scene animation (use threejs-builder skill). This agent integrates Rive skeletal animation, state machines, and character pipelines. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | | sqlite-peewee-engineer | SQLite with Peewee ORM: model definition, query optimization, migrations, transactions | peewee, sqlite, ORM, python database, playhouse | | server databases like Postgres/MySQL (use database-engineer); general Python features or debugging outside the data layer (use python-general-engineer) | python-general-engineer, database-engineer | -| swift-general-engineer | Swift development: iOS, macOS, server-side Swift, SwiftUI, concurrency, testing | programming, ios, macos, xcode, programmingui, uikit, appkit, watchos, tvos, visionos, vapor, spm, programming-package-manager, programminglint, programmingformat, xctest, programming actor, programming sendable, programming-combine, programmingdata, coredata | | Swift concurrency patterns or test-only work in isolation (use programming skill) — this agent handles full Swift development | workflow, testing, review | -| system-upgrade-engineer | Systematic toolkit upgrades: adapt agents, skills, hooks when Claude Code ships updates | upgrade agents, system upgrade, claude update, upgrade skills, adapt workflow, apply claude update, apply update, system health, update system, new claude version, apply process | | editing one skill, routing entry, or ADR (use toolkit-governance-engineer); regenerating routing INDEX files (use toolkit skill); writing a new Python hook implementation (use hook-development-engineer); explaining which agent or workflow to run (use workflow-help skill). This agent adapts agents, skills, and hooks across the fleet when Claude Code ships a release. | toolkit, toolkit, assessment, toolkit, pr-workflow | -| technical-documentation-engineer | Technical documentation: API docs, system architecture, runbooks, enterprise standards | API documentation, technical docs, documentation validation, integration guide | | explainers, opinion pieces, or long-form articles for a public audience (use technical-journalist-writer); detecting drift between existing docs and code (use docs-sync-checker skill); building a local documentation MCP server (use mcp-local-docs-engineer). This agent writes API references, architecture docs, runbooks, and integration guides. | testing | -| technical-journalist-writer | Technical journalism: explainers, opinion pieces, analysis articles, long-form content | technical article, technical writing, journalist voice, technical writer, technical journalism | | API references, runbooks, or architecture documentation (use technical-documentation-engineer); drafting through a stored voice profile (use writing skill); building or validating a voice profile (use writing skill); SEO, campaign, and launch copy (use content skill). This agent writes technical journalism: explainers, opinion, and analysis. | writing | -| testing-automation-engineer | Testing automation: Vitest, Playwright, E2E, coverage enforcement, CI/CD integration | testing, E2E, playwright, vitest, test automation, visual regression | | Playwright-only E2E test authoring (use testing skill) — this agent covers full test automation strategy including Vitest, coverage, and CI integration | testing, testing | -| toolkit-governance-engineer | Toolkit governance: edit skills, update routing tables, manage ADR lifecycle, enforce standards | edit skill, routing table governance, ADR management, toolkit maintenance, check coverage, skill compliance, hook standardization, cross-component consistency, create skill, new skill, scaffold skill, build a skill, create a skill | | authoring one new skill end-to-end (use toolkit skill); mechanically regenerating routing INDEX files (use the toolkit skill); adapting the fleet to a new Claude Code release (use system-upgrade-engineer); writing a Python hook implementation (use hook-development-engineer); scaffolding a new multi-component pipeline (use pipeline-orchestrator-engineer). This agent governs fleet-wide policy, routing consistency, and ADR conformance, not single-skill scaffolding or index sync. | assessment, toolkit, docs-sync-checker | -| typescript-debugging-engineer | TypeScript debugging: race conditions, async/await issues, type errors, runtime exceptions | typescript debug, async bug, race condition, type error, production error, memory leak | | building components, state management, or build configuration (use typescript-frontend-engineer); running tsc and clearing reported type errors mechanically (use programming skill); React Native runtime issues (use react-native-engineer); Core Web Vitals and bundle performance (use performance-optimization-engineer). This agent diagnoses TypeScript race conditions, async bugs, type failures, and production runtime exceptions. | workflow, typescript-frontend-engineer | -| typescript-frontend-engineer | TypeScript frontend architecture: type-safe components, state management, build optimization | typescript, react, next.js, frontend, .tsx, .ts, zod | | React Native mobile apps (use react-native-engineer); diagnosing race conditions, async bugs, or production runtime exceptions (use typescript-debugging-engineer); e-commerce carts, Stripe, and checkout flows (use nextjs-ecommerce-engineer); portfolio and gallery sites (use react-portfolio-engineer). This agent builds TypeScript frontend architecture: type-safe components, state, and build configuration. | code-quality, programming | -| ui-design-engineer | UI/UX frontend: frontend systems, responsive layouts, accessibility, animations | UI, frontend, tailwind, accessibility, responsive, animations, frontend system | | non-visual frontend: API, schema, or system frontend (use the matching engineer); writing frontend docs (use a workflow skill); frontend methodology, user-research analysis, or accessibility audit documentation (use frontend skill); combat visual effects and animation juice (use combat-effects-upgrade). This agent is visual UI/UX implementation only. | frontend, typescript-frontend-engineer | +| swift-general-engineer | Swift development: iOS, macOS, server-side Swift, SwiftUI, concurrency, testing | swift, ios, macos, xcode, swiftui, uikit, appkit, watchos, tvos, visionos, vapor, spm, swift-package-manager, swiftlint, swiftformat, xctest, swift actor, swift sendable, swift-combine, swiftdata, coredata | | Swift concurrency patterns or test-only work in isolation (use swift skill) — this agent handles full Swift development | workflow, verification-before-completion, systematic-code-review | +| system-upgrade-engineer | Systematic toolkit upgrades: adapt agents, skills, hooks when Claude Code ships updates | upgrade agents, system upgrade, claude update, upgrade skills, adapt workflow, apply claude update, apply update, system health, update system, new claude version, apply retro | | editing one skill, routing entry, or ADR (use toolkit-governance-engineer); regenerating routing INDEX files (use routing-table-updater skill); writing a new Python hook implementation (use hook-development-engineer); explaining which agent or workflow to run (use workflow-help skill). This agent adapts agents, skills, and hooks across the fleet when Claude Code ships a release. | toolkit-evolution, agent-evaluation, codebase-overview, routing-table-updater, pr-workflow | +| technical-documentation-engineer | Technical documentation: API docs, system architecture, runbooks, enterprise standards | API documentation, technical docs, documentation validation, integration guide | | explainers, opinion pieces, or long-form articles for a public audience (use technical-journalist-writer); detecting drift between existing docs and code (use docs-sync-checker skill); building a local documentation MCP server (use mcp-local-docs-engineer). This agent writes API references, architecture docs, runbooks, and integration guides. | verification-before-completion | +| technical-journalist-writer | Technical journalism: explainers, opinion pieces, analysis articles, long-form content | technical article, technical writing, journalist voice, technical writer, technical journalism | | API references, runbooks, or architecture documentation (use technical-documentation-engineer); drafting through a stored voice profile (use voice-writer skill); building or validating a voice profile (use create-voice or voice-validator skills); SEO, campaign, and launch copy (use marketing skill). This agent writes technical journalism: explainers, opinion, and analysis. | voice-writer | +| testing-automation-engineer | Testing automation: Vitest, Playwright, E2E, coverage enforcement, CI/CD integration | testing, E2E, playwright, vitest, test automation, visual regression | | Playwright-only E2E test authoring (use e2e-testing skill) — this agent covers full test automation strategy including Vitest, coverage, and CI integration | test-driven-development, e2e-testing | +| toolkit-governance-engineer | Toolkit governance: edit skills, update routing tables, manage ADR lifecycle, enforce standards | edit skill, routing table governance, ADR management, toolkit maintenance, check coverage, skill compliance, hook standardization, cross-component consistency, create skill, new skill, scaffold skill, build a skill, create a skill | | authoring one new skill end-to-end (use skill-creator); mechanically regenerating routing INDEX files (use the routing-table-updater skill); adapting the fleet to a new Claude Code release (use system-upgrade-engineer); writing a Python hook implementation (use hook-development-engineer); scaffolding a new multi-component pipeline (use pipeline-orchestrator-engineer). This agent governs fleet-wide policy, routing consistency, and ADR conformance, not single-skill scaffolding or index sync. | adr-consultation, routing-table-updater, docs-sync-checker, skill-creator | +| typescript-debugging-engineer | TypeScript debugging: race conditions, async/await issues, type errors, runtime exceptions | typescript debug, async bug, race condition, type error, production error, memory leak | | building components, state management, or build configuration (use typescript-frontend-engineer); running tsc and clearing reported type errors mechanically (use typescript-check skill); React Native runtime issues (use react-native-engineer); Core Web Vitals and bundle performance (use performance-optimization-engineer). This agent diagnoses TypeScript race conditions, async bugs, type failures, and production runtime exceptions. | workflow, typescript-frontend-engineer | +| typescript-frontend-engineer | TypeScript frontend architecture: type-safe components, state management, build optimization | typescript, react, next.js, frontend, .tsx, .ts, zod | | React Native mobile apps (use react-native-engineer); diagnosing race conditions, async bugs, or production runtime exceptions (use typescript-debugging-engineer); e-commerce carts, Stripe, and checkout flows (use nextjs-ecommerce-engineer); portfolio and gallery sites (use react-portfolio-engineer). This agent builds TypeScript frontend architecture: type-safe components, state, and build configuration. | universal-quality-gate, typescript-check | +| ui-design-engineer | UI/UX design: design systems, responsive layouts, accessibility, animations | UI, design, tailwind, accessibility, responsive, animations, design system | | non-visual design: API, schema, or system design (use the matching engineer); writing design docs (use a planning skill); design methodology, user-research analysis, or accessibility audit documentation (use design skill); combat visual effects and animation juice (use combat-effects-upgrade). This agent is visual UI/UX implementation only. | distinctive-frontend-design, typescript-frontend-engineer | -## SKILLS (44) +## SKILLS (43) | name | when to call (description) | triggers | FORCE | not_for | pairs_with | |---|---|---|---|---|---| -| accounts | Manage multiple Claude Code accounts: add, list, check, launch, and install shell aliases for 10+ isolated CLAUDE_CONFIG_DIR profiles. | multiple accounts, multiple claude accounts, CLAUDE_CONFIG_DIR, claude account, switch accounts, account alias, claude login, c1 c2 aliases, claude max account, manage accounts, add claude account | | API key rotation (use secrets management), Anthropic billing (use the web console) | deploy | | architecture-deepening | Improve architecture across modules by deepening interfaces. | improve architecture, improve codebase architecture, improve the codebase architecture, find architecture improvements, deepen architecture, find shallow modules, architecture improvement, module depth analysis, deepening opportunities, improve module interfaces, architecture deepening | | local cleanup/refactoring (workflow or planning), feature design (feature-lifecycle), architecture overview/explanation (codebase-overview), or vague complexity reduction; requires cross-module interface or caller-burden evidence | review, assessment | | assessment | Assessment: read-only inspection, codebase overview, value analysis, health checks, ADR consultation, decision analysis, multi-perspective critique. | inspect without changing, read-only, audit current state, onboard to codebase, codebase structure, give me an overview, summarize this repo, repo value analysis, what can we learn from, service status, check health, is service running, validate endpoints, consult on ADR, architecture consultation, adr consultation, help me decide, decision matrix, pros and cons, trade-offs, critique these ideas, devil's advocate, stress test proposals, roast this, poke holes in this | | code review with findings (use review), building or fixing (use workflow) | review, workflow, security | | auto-dream | Background memory consolidation — overnight review, merge, and injection payload for memory files. | dream, consolidate memories, clean up memories, memory maintenance, memory consolidation, deduplicate memories | | | | @@ -77,13 +77,13 @@ CI check: `python3 scripts/generate-routing-map.py --check`. | do | Classify user requests and route to the correct agent + skill. | route task, classify request, which agent, delegate to skill, smart router | | | | | docs-sync-checker | Detect documentation drift against filesystem state. | check doc drift, sync documentation, stale docs, documentation drift, README outdated | | | toolkit, assessment | | domain | Domain-specific: SAP Commerce, OpenSearch detection, WordPress validation, enterprise search. | sapcc review, sapcc audit, sapcc compliance, sapcc standards, siem detection, sigma rule, mitre att&ck mapping, detection engineering, opensearch detection, anomaly detection rule, soc escalation, validate wordpress post, check live post, wordpress post validation, enterprise search, search relevance, search ranking, BM25, query understanding, search quality, OpenSearch, Elasticsearch, vector search, hybrid search, search tuning | | general code review (use review), general security (use security) | golang-general-engineer, opensearch-elasticsearch-engineer, programming | -| frontend | Frontend: UI design, distinctive visual styles, HTML artifacts, Three.js 3D. | design, UX copy, design system, accessibility, WCAG, design handoff, frontend design, visual identity, HTML artifact, make HTML, self-contained HTML, threejs, three.js, 3D web, 3D scene, WebGL, WebGPU, react three fiber, r3f, text animation, card effects, pptx, powerpoint, slide deck, design critique | | game UI (use game-dev), backend work (use workflow), brand or logo design without code | typescript-frontend-engineer, ui-design-engineer, game-dev | +| frontend | Frontend: UI design, distinctive visual styles, HTML artifacts, Three.js 3D. | design, UX copy, design system, accessibility, WCAG, design handoff, frontend design, visual identity, HTML artifact, make HTML, self-contained HTML, threejs, three.js, 3D web, 3D scene, WebGL, WebGPU, react three fiber, r3f, text animation, card effects, pptx, powerpoint, slide deck, design critique | | game UI (use game-dev), backend work (use workflow), brand or logo design without code | typescript-frontend-engineer, ui-frontend-engineer, game-dev | | game-dev | Game development: design, asset pipelines, Phaser, vanilla JS frontend, GM mode, match booking. | game design, game design audit, game improvement, core loop, game feel, game balance, game economy, make game, game pipeline, game audio, deploy game, phaser, 2d game, platformer, arcade physics, tilemap, AI sprite, generate sprite, spritesheet pipeline, animated spritesheet, large GM implementation, GM systems overhaul, pixel art, screen shake, game qa | yes | non-game frontend (use frontend), non-game backend (use workflow) | frontend, workflow | | github | GitHub: notification triage, profile rule extraction. | github notifications, triage notifications, check notifications, notification cleanup, github inbox, triage issues, triage pull requests, pr queue, github profile rules, profile conventions | | git commit/push/PR operations (use pr-workflow) | | | grill-jev | Broad Jev interrogation: generate up to 50 context-specific questions about a plan, spec, design, code artifact, or any topic — using all primitive... | grill, validate plan, validate spec, validate design, interrogate, stress test plan, stress test spec, 50 questions, deep validation, grill this, grill the plan, jev interrogation, sanity check plan, sanity check spec, check my plan, check my design, plan validation, spec validation, design validation, architecture review jev, verify plan, verify spec, challenge plan, challenge design, find holes, find gaps in plan | yes | Writing new Jev programs (use building-with-jev). Routing requests (use do). Code review for quality/style (use review). Security scanning (use security). This skill runs a broad interrogation battery against a supplied artifact — it does not build Jev programs. | building-with-jev, review | | image-gen | AI image generation: Gemini and Nano Banana backends; single/series/batch workflows with prompt-to-disk. | generate image, create image, image generation, AI image, gemini image, make image, draw, illustrate, sprite generation, card art, batch generation, series of images, character art, pixel art, image post-processing, make a picture, create art, generate a picture, make art, generate artwork, create artwork | | HTML visualization or charts (use html-artifact), or deterministic non-AI palette/matrix pixel art (use game-asset-generator) | python-general-engineer, game-dev | | integration-checker | Verify cross-component wiring and data flow. | integration check, check integration, verify wiring, are components connected, check connections, integration-checker, wiring check | | | workflow, review | -| jev-design | Design React interfaces with the full shadcn/ui catalog. | jev design, design with jev, shadcn component plan, choose shadcn components, shadcn ui design, component style recipe, which shadcn component, shadcn theme recipe, pick shadcn components, design a shadcn screen | | | frontend, building-with-jev | +| jev-design | Plan React interfaces with shadcn components and style recipes selected from bounded catalogs by live Jev. | jev design, design with jev, shadcn component plan, choose shadcn components, shadcn ui design, component style recipe | | | frontend, building-with-jev | | joy-check | Validate content framing on joy-grievance spectrum. | joy check, check framing, tone check, negative framing, joy validation, too negative, reframe positively, positive framing check, instruction framing | | | writing, toolkit | | kubernetes | Kubernetes operations: debugging, security, RBAC, and infrastructure tooling. | kubernetes debug, pod failure, pod crashloop, kubectl logs, OOMKilled, pod pending, kubernetes security, k8s RBAC, RBAC setup, pod security policy, network policy, cobalt core, cobaltcore, kvm-exporter, kvm exporter, hypervisor metrics, libvirt exporter, cloud hypervisor | | | assessment, programming, prometheus-grafana-engineer | | markdown-converter | Convert PDF, Office, HTML, data, media, ZIP to Markdown. | convert to markdown, markitdown, extract text from PDF, PDF to markdown, docx to markdown, ingest document, read this PDF, read this document, extract text from document, convert PDF, convert document, pptx to markdown, xlsx to markdown | | | research, domain | @@ -93,14 +93,14 @@ CI check: `python3 scripts/generate-routing-map.py --check`. | programming | Language-specific patterns and tooling: Go, Kotlin, PHP, Swift, TypeScript. | .go, go test, *_test.go, goroutine, sync.Mutex, context.Context, Go code review, Go lint, sapcc, make check, kotlin coroutines, kotlin Flow, suspend function, kotlin testing, kotest, php quality, php code review, PSR standards, phpstan, phpunit, swift concurrency, swift async await, Swift Actor, XCTest, TypeScript check, tsc noEmit, tsc errors | yes | Python (agents handle directly), general debugging (use debugging). When a dedicated language skill exists (php, swift, kotlin), prefer it over this umbrella | golang-general-engineer, kotlin-general-engineer, php-general-engineer, swift-general-engineer, typescript-frontend-engineer | | quick | Tracked lightweight execution with composable rigor flags: --trivial, --discuss, --research, --full. | quick task, small change, ad hoc task, add a flag, small refactor, targeted fix, quick fix, typo fix, fix typo, fix the typo, one-line change, trivial fix, rename variable, rename this variable, update value, fix import, small mistake, small mistake in, mistake in spelling, spelling mistake, spelling fix, fix the spelling, typo in, small fix in, small fix, tiny fix | yes | 'quick' as speed preference, general bug diagnosis requiring investigation | | | research | Research: structured investigation, fact-checking, explanation traces. | research-pipeline, research, formal research, research with artifacts, systematic investigation, research report, gather evidence, fact check, fact-check, verify claims, check facts, verify this quote, is this accurate, is this true, check this claim, verify this, are these numbers right, why did you, explain routing, show trace, decision log, why that agent, explain decision, show decisions, trace log | yes | code review (use review), security audit (use security) | review, writing | -| review | Code review: systematic single-file, parallel multi-reviewer, full-repo audit, PR diff review. | review code, code review, code review methodology, structured review, code audit, review methodology, comprehensive review, parallel review, 3-reviewer review, multi-reviewer, concurrent review, full repo review, review entire repo, codebase health check, review all files, full codebase review, audit the codebase, codebase audit, review whole repo, sweep all source files, repo health, review this PR, review my PR, review my changes, PR review, diff review | yes | security scanning (use security), linting (use code-quality), git push/commit/PR (use pr-workflow) | security, testing | -| security | Security: review git changes for vulnerabilities, threat-model a system's attack surface, audit supply-chain risks. | security review, review my changes for security, review for security, security scan, review for vulnerabilities, scan for vulnerabilities, check for security issues, security issues, audit for vulnerabilities, audit auth, audit vulnerabilities, threat model, security audit, supply chain scan, deny list, security posture, injection scan, surface scan, audit hooks, audit skills | yes | general code review (use review), non-security quality checks (use code-quality) — only for security-specific work: vulnerability scanning, threat modeling, supply-chain audits, auth reviews | review, reviewer-system | -| testing | Testing: TDD, E2E, preferred patterns, test-value audits, verification, agent testing. | TDD, test first, red green refactor, write tests first, test-driven, tests before code, flaky test, brittle test, test smell, test quality issue, slow tests, over-mocking, test audit, audit tests, low-value tests, prune tests, test bloat, delete useless tests, test agents, agent testing, subagent testing, run vitest, JavaScript tests, TypeScript tests, playwright, E2E test, end-to-end, browser test, verify completion, run tests, final verification | | code review (use review), linting (use code-quality) | review, code-quality, workflow | -| toolkit | Toolkit management: create and evaluate skills and agents, manage routing tables, generate Claude.md. | create skill, create agent, scaffold skill, scaffold agent, new skill, new agent, skill template, agent template, eval skill, evaluate agent, benchmark skill, benchmark agents, compare agents, A/B test agents, optimize description, evolve toolkit, toolkit evolution, self-improve, update routing tables, routing maintenance, generate claude.md, create claude.md, compose skills, bake-off, skill quality, weaker model, weak model uplift, improve skill for weaker models | yes | application code (use workflow), code review (use review) | review, workflow | +| review | Code review: systematic single-file, parallel multi-reviewer, full-repo audit, PR diff review. | review code, code review, code review methodology, structured review, code audit, review methodology, comprehensive review, parallel review, 3-reviewer review, multi-reviewer, concurrent review, full repo review, review entire repo, codebase health check, review all files, full codebase review, audit the codebase, codebase audit, review whole repo, sweep all source files, repo health, review this PR, review my PR, PR review, diff review | yes | security scanning (use security), linting (use code-quality), git push/commit/PR (use pr-workflow) | security, testing | +| security | Security: review git changes for vulnerabilities, threat-model a system's attack surface, audit supply-chain risks. | security review, review my changes for security, review my changes, review for security, security scan, review for vulnerabilities, scan for vulnerabilities, check for security issues, security issues, audit for vulnerabilities, audit auth, audit vulnerabilities, threat model, security audit, supply chain scan, deny list, security posture, injection scan, surface scan, audit hooks, audit skills | yes | general code review (use review), non-security quality checks (use code-quality) — only for security-specific work: vulnerability scanning, threat modeling, supply-chain audits, auth reviews | review, reviewer-system | +| testing | Testing: TDD, E2E, preferred patterns, verification, agent testing. | TDD, test first, red green refactor, write tests first, test-driven, tests before code, flaky test, brittle test, test smell, test quality issue, slow tests, over-mocking, test agents, agent testing, subagent testing, run vitest, JavaScript tests, TypeScript tests, playwright, E2E test, end-to-end, browser test, verify completion, run tests, final verification | | code review (use review), linting (use code-quality) | review, code-quality, workflow | +| toolkit | Toolkit management: create and evaluate skills and agents, manage routing tables, generate Claude.md. | create skill, create agent, scaffold skill, scaffold agent, new skill, new agent, skill template, agent template, eval skill, evaluate agent, benchmark skill, benchmark agents, compare agents, A/B test agents, optimize description, evolve toolkit, toolkit evolution, self-improve, update routing tables, routing maintenance, generate claude.md, create claude.md, compose skills, bake-off, skill quality | yes | application code (use workflow), code review (use review) | review, workflow | | video-editing | Video editing pipeline: cut footage, assemble clips via FFmpeg and Remotion. | edit video, cut footage, make vlog, screen recording, video workflow, ffmpeg, remotion, demo video, make a clip, assemble clips, video editing, image to video, audio visualization, static video, mp4 from image, music video, podcast video, video from image, combine image audio, album art video, cover art video, video transcript, youtube transcript, extract transcript, download subtitles, what does this video say | | | typescript-frontend-engineer, research | | workflow | Structured work: multi-phase tasks, feature builds, planning, objective loops, hill climbing. | workflow, multi-phase task, feature design, feature plan, feature implement, build feature end to end, full feature lifecycle, write spec, define requirements, create plan, create tasks, keep working until, iterate until done, drive this to done, make this faster, speed this up, reduce latency, profile and optimize, hill climb on this metric, tidy up, clean up, untangle, reorganize, structured pipeline, phased execution | yes | code review (use review), testing (use testing), security (use security) | review, testing, security, pr-workflow | | workflow-help | Interactive guide to workflow system: agents, skills, routing, execution patterns. | how does routing work, what skills exist, system help, explain workflow, I am stuck, toolkit help | yes | | workflow, do | -| writing | Writing: voice creation and validation, prose editing, anti-AI cleanup, professional communication, translation. | write article, blog post, write in voice, voice pipeline, voice writer, validate voice, voice fidelity, write email, draft memo, executive summary, status update, meeting notes, pushback email, disagree professionally, translate, translation, localize, draft article, generate content, professional communication, draft response | yes | code comments (use code-quality), API documentation (use docs-sync-checker), scheduling or calendaring content (use content), code review (use review) | joy-check, content | +| writing | Writing: voice creation and validation, prose editing, anti-AI cleanup, professional communication, translation. | write article, blog post, write in voice, voice pipeline, voice-writer, voice writer, validate voice, voice fidelity, write email, draft memo, executive summary, status update, meeting notes, pushback email, disagree professionally, translate, translation, localize, draft article, generate content, professional communication, draft response | yes | code comments (use code-quality), API documentation (use docs-sync-checker), scheduling or calendaring content (use content), code review (use review) | joy-check, content | ## PIPELINES (25) From 94bc68d4fdf6b469370741dd70a82e60e86029ac Mon Sep 17 00:00:00 2001 From: thecodingshrimp Date: Wed, 30 Sep 2026 17:57:59 +0200 Subject: [PATCH 5/6] fix routing-map.md stale after cherry-pick conflict resolution: regenerate to drop phantom agents (golang-compact, github-profile-rules-engineer) and sync agent count to 42 --- docs/routing-map.md | 98 ++++++++++++++++++++++----------------------- 1 file changed, 49 insertions(+), 49 deletions(-) diff --git a/docs/routing-map.md b/docs/routing-map.md index 0f66e959e..a9973aed2 100644 --- a/docs/routing-map.md +++ b/docs/routing-map.md @@ -6,58 +6,58 @@ Single source of truth: agent/skill frontmatter and `pipeline-index.json`. Regenerate: `python3 scripts/generate-routing-map.py`. CI check: `python3 scripts/generate-routing-map.py --check`. -## AGENTS (43) +## AGENTS (42) | name | when to call (description) | triggers | FORCE | not_for | pairs_with | |---|---|---|---|---|---| -| ansible-automation-engineer | Ansible automation: playbooks, roles, collections, Molecule testing, Vault security | ansible, playbook, automation, molecule, ansible-tower, AWX | | Kubernetes manifests, Helm charts, or cluster deploys (use kubernetes-helm-engineer); Claude Code hook or toolkit automation (use hook-development-engineer); application test automation (use testing-automation-engineer). This agent writes Ansible playbooks, roles, collections, and Molecule tests. | verification-before-completion, kubernetes-helm-engineer | -| combat-effects-upgrade | Zero-dependency combat visual upgrades: CSS particle replacement, Framer Motion combat juice, CSS 3D card transforms | combat effects, CSS particles, particle replacement, framer motion combat, combat animations, card transforms, CSS 3D, combat juice, visual effects upgrade, effects.ts, combat polish | | WebGL or GPU-particle combat rendering (use pixijs-combat-renderer); skeletal or character animation rigs (use rive-skeletal-animator); general design-system, layout, or accessibility work (use ui-design-engineer). This agent does zero-dependency CSS and Framer Motion combat effects. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | -| data-engineer | Data pipelines, ETL/ELT, warehouse design, dimensional modeling, stream processing | data pipeline, ETL, ELT, dbt, Airflow, Prefect, Dagster, dimensional model, data warehouse, star schema, snowflake schema, data lake, data quality, streaming, Kafka, Spark, Flink, BigQuery, Redshift, Parquet, Delta Lake, Iceberg, data vault, slowly changing dimension, SCD, data lineage | | OLTP schema/query tuning (use database-engineer); RabbitMQ or AMQP broker topology, clustering, and high availability (use rabbitmq-messaging-engineer) | database-engineer, data-analysis | -| database-engineer | Database design, optimization, query performance, migrations, indexing strategies | database, schema, SQL, postgres, mysql, indexing, query optimization | | embedded SQLite or Peewee ORM work (use sqlite-peewee-engineer) — this agent's foreign-key and zero-downtime migration mandates assume a server database and misfire on embedded SQLite; analytics warehouses, ETL/ELT, dimensional models, or stream processing (use data-engineer); OpenSearch or Elasticsearch cluster tuning (use opensearch-elasticsearch-engineer). This agent designs and tunes server databases such as Postgres and MySQL. | nodejs-api-engineer, sqlite-peewee-engineer, data-engineer | -| github-profile-rules-engineer | Extract coding conventions and style rules from GitHub user profiles via API | github rules, profile analysis, coding style extraction, github conventions, programming rules | | deriving conventions from a local repository (use codebase-overview skill); generating a project CLAUDE.md from repo analysis (use generate-claudemd skill); building a writing-voice profile from author samples (use create-voice skill); reviewing code against conventions already adopted (use reviewer-code). This agent extracts coding rules from GitHub user profiles via the API. | codebase-overview, generate-claudemd | -| golang-general-engineer | Go development: features, debugging, code review, performance | go, golang, .go files, gofmt, go mod, goroutine, channel, gopls | | tasks using 'go' as a verb (go ahead, go fix this); Kotlin coroutine work (use kotlin-general-engineer); Go concurrency patterns in isolation (use go-patterns skill) | go-patterns | -| golang-general-engineer-compact | Compact Go development for tight context budgets | tight context, compact, focused go | | general Go feature/debug work via /do (use golang-general-engineer) — only tight-context budgets and per-package review fan-out | go-patterns | -| hook-development-engineer | Python hook development for Claude Code event-driven system and learning database | create hook, hook development, event handler, PostToolUse, PreToolUse, SessionStart, learning database, error detection hook | | auditing hook conventions, registration, or timeouts fleet-wide (use toolkit-governance-engineer); adapting existing hooks to a new Claude Code release (use system-upgrade-engineer); querying or graduating captured learnings (use learn skill); general Python features outside the hook system (use python-general-engineer). This agent writes and debugs Python hook implementations. | verification-before-completion, python-quality-gate | -| kotlin-general-engineer | Kotlin development: features, coroutines, debugging, code quality, multiplatform | kotlin, ktor, koin, coroutine, suspend fun, kotlin flow, StateFlow, kotest, mockk, gradle-kts, detekt, ktlint, ktfmt, android kotlin, kotlin-multiplatform | | writing Kotlin tests (use kotlin); coroutine or Flow patterns in isolation (use kotlin); Go goroutine work (use golang-general-engineer). This agent writes and debugs Kotlin features. | workflow, verification-before-completion, systematic-code-review | -| kubernetes-helm-engineer | Kubernetes and Helm: deployments, troubleshooting, cloud-native infrastructure | kubernetes, helm, k8s, kubectl, statefulset, argocd, deployment | | Ansible playbooks and host configuration management (use ansible-automation-engineer); Prometheus or Grafana monitoring, alerting, and PromQL (use prometheus-grafana-engineer); Perses dashboards, plugins, or operator work (use perses-engineer); cluster debugging, RBAC, and security operations (use kubernetes skill). This agent authors and fixes Kubernetes manifests and Helm charts. | verification-before-completion, prometheus-grafana-engineer | -| mcp-local-docs-engineer | MCP server development for local documentation access in TypeScript/Node | MCP, docs server, documentation server, hugo | | using existing MCP tools/servers (only building local-docs MCP servers); web performance or LCP optimization (use performance-optimization-engineer); writing the documentation content itself, API references, or runbooks (use technical-documentation-engineer) | verification-before-completion | -| nextjs-ecommerce-engineer | Use this agent when building a NextJS e-commerce site: shopping cart, Stripe payments, product catalogs, order management, and checkout flows | next.js e-commerce, nextjs ecommerce, shopping cart, stripe, e-commerce, online store, product catalog | | general React or Next.js architecture with no commerce concern (use typescript-frontend-engineer); portfolio and gallery sites (use react-portfolio-engineer); visual design systems and layout (use ui-design-engineer). This agent builds Next.js commerce: carts, Stripe checkout, catalogs, and orders. | verification-before-completion, typescript-frontend-engineer | -| nodejs-api-engineer | Use this agent when you need expert assistance with NodeJS backend API development: REST endpoints, authentication, file uploads, webhooks, middleware, and database integration | node.js, nodejs, express, API, backend, webhook, authentication | | KPI dashboards or data analysis (use data-analysis skill) — this agent builds Node.js REST/GraphQL APIs | systematic-code-review, database-engineer | -| opensearch-elasticsearch-engineer | OpenSearch/Elasticsearch: cluster management, performance tuning, index optimization | opensearch, elasticsearch, search cluster, logstash, kibana, search performance | | search-relevance UX or query-DSL authoring inside an app (use enterprise-search); relational database indexing and query tuning (use database-engineer). This agent runs and tunes the OpenSearch or Elasticsearch cluster itself. | verification-before-completion, enterprise-search | -| performance-optimization-engineer | Web performance optimization: Core Web Vitals, rendering, bundle analysis, monitoring | core web vitals, LCP, bundle size, render performance, web performance, lighthouse | | backend/algorithmic performance in a specific language (use that language engineer); MCP server development (use mcp-local-docs-engineer) — only web/frontend runtime and load performance | verification-before-completion | +| ansible-automation-engineer | Ansible automation: playbooks, roles, collections, Molecule testing, Vault security | ansible, playbook, automation, molecule, ansible-tower, AWX | | Kubernetes manifests, Helm charts, or cluster deploys (use kubernetes-helm-engineer); Claude Code hook or toolkit automation (use hook-development-engineer); application test automation (use testing-automation-engineer). This agent writes Ansible playbooks, roles, collections, and Molecule tests. | testing, kubernetes-helm-engineer | +| combat-effects-upgrade | Zero-dependency combat visual upgrades: CSS particle replacement, Framer Motion combat juice, CSS 3D card transforms | combat effects, CSS particles, particle replacement, framer motion combat, combat animations, card transforms, CSS 3D, combat juice, visual effects upgrade, effects.ts, combat polish | | WebGL or GPU-particle combat rendering (use pixijs-combat-renderer); skeletal or character animation rigs (use rive-skeletal-animator); general frontend-system, layout, or accessibility work (use ui-design-engineer). This agent does zero-dependency CSS and Framer Motion combat effects. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | +| data-engineer | Data pipelines, ETL/ELT, warehouse frontend, dimensional modeling, stream processing | data pipeline, ETL, ELT, dbt, Airflow, Prefect, Dagster, dimensional model, data warehouse, star schema, snowflake schema, data lake, data quality, streaming, Kafka, Spark, Flink, BigQuery, Redshift, Parquet, Delta Lake, Iceberg, data vault, slowly changing dimension, SCD, data lineage | | OLTP schema/query tuning (use database-engineer); RabbitMQ or AMQP broker topology, clustering, and high availability (use rabbitmq-messaging-engineer) | database-engineer, data | +| database-engineer | Database frontend, optimization, query performance, migrations, indexing strategies | database, schema, SQL, postgres, mysql, indexing, query optimization | | embedded SQLite or Peewee ORM work (use sqlite-peewee-engineer) — this agent's foreign-key and zero-downtime migration mandates assume a server database and misfire on embedded SQLite; analytics warehouses, ETL/ELT, dimensional models, or stream processing (use data-engineer); OpenSearch or Elasticsearch cluster tuning (use opensearch-elasticsearch-engineer). This agent frontends and tunes server databases such as Postgres and MySQL. | nodejs-api-engineer, sqlite-peewee-engineer, data-engineer | +| github-engineer | Extract coding conventions and style rules from GitHub user profiles via API | github rules, profile analysis, coding style extraction, github conventions, programming rules | | deriving conventions from a local repository (use assessment skill); generating a project CLAUDE.md from repo analysis (use toolkit skill); building a writing-voice profile from author samples (use writing skill); reviewing code against conventions already adopted (use reviewer-code). This agent extracts coding rules from GitHub user profiles via the API. | assessment, toolkit | +| golang-general-engineer | Go development: features, debugging, code review, performance | go, golang, .go files, gofmt, go mod, goroutine, channel, gopls | | tasks using 'go' as a verb (go ahead, go fix this); Kotlin coroutine work (use kotlin-general-engineer); Go concurrency patterns in isolation (use programming skill) | programming | +| hook-development-engineer | Python hook development for the Claude Code event-driven system and its telemetry store | create hook, hook development, event handler, PostToolUse, PreToolUse, SessionStart, routing telemetry hook, hook registration | | auditing hook conventions, registration, or timeouts fleet-wide (use toolkit-governance-engineer); adapting existing hooks to a new Claude Code release (use system-upgrade-engineer); reading routing telemetry that hooks already recorded (use the process skill); general Python features outside the hook system (use python-general-engineer). This agent writes and debugs Python hook implementations. | testing, code-quality | +| kotlin-general-engineer | Kotlin development: features, coroutines, debugging, code quality, multiplatform | programming, ktor, koin, coroutine, suspend fun, programming flow, StateFlow, kotest, mockk, gradle-kts, detekt, ktlint, ktfmt, android programming, programming-multiplatform | | writing Kotlin tests (use programming); coroutine or Flow patterns in isolation (use programming); Go goroutine work (use golang-general-engineer). This agent writes and debugs Kotlin features. | workflow, testing, review | +| kubernetes-helm-engineer | Kubernetes and Helm: deployments, troubleshooting, cloud-native infrastructure | kubernetes, helm, k8s, kubectl, statefulset, argocd, deployment | | Ansible playbooks and host configuration management (use ansible-automation-engineer); Prometheus or Grafana monitoring, alerting, and PromQL (use prometheus-grafana-engineer); Perses dashboards, plugins, or operator work (use perses-engineer); cluster debugging, RBAC, and security operations (use kubernetes skill). This agent authors and fixes Kubernetes manifests and Helm charts. | testing, prometheus-grafana-engineer | +| mcp-local-docs-engineer | MCP server development for local documentation access in TypeScript/Node | MCP, docs server, documentation server, hugo | | using existing MCP tools/servers (only building local-docs MCP servers); web performance or LCP optimization (use performance-optimization-engineer); writing the documentation content itself, API references, or runbooks (use technical-documentation-engineer) | testing | +| nextjs-ecommerce-engineer | Use this agent when building a NextJS e-commerce site: shopping cart, Stripe payments, product catalogs, order management, and checkout flows | next.js e-commerce, nextjs ecommerce, shopping cart, stripe, e-commerce, online store, product catalog | | general React or Next.js architecture with no commerce concern (use typescript-frontend-engineer); portfolio and gallery sites (use react-portfolio-engineer); visual frontend systems and layout (use ui-design-engineer). This agent builds Next.js commerce: carts, Stripe checkout, catalogs, and orders. | testing, typescript-frontend-engineer | +| nodejs-api-engineer | Use this agent when you need expert assistance with NodeJS backend API development: REST endpoints, authentication, file uploads, webhooks, middleware, and database integration | node.js, nodejs, express, API, backend, webhook, authentication | | KPI dashboards or data analysis (use data skill) — this agent builds Node.js REST/GraphQL APIs | review, database-engineer | +| opensearch-elasticsearch-engineer | OpenSearch/Elasticsearch: cluster management, performance tuning, index optimization | opensearch, elasticsearch, search cluster, logstash, kibana, search performance | | search-relevance UX or query-DSL authoring inside an app (use domain skill); relational database indexing and query tuning (use database-engineer). This agent runs and tunes the OpenSearch or Elasticsearch cluster itself. | testing, domain | +| performance-optimization-engineer | Web performance optimization: Core Web Vitals, rendering, bundle analysis, monitoring | core web vitals, LCP, bundle size, render performance, web performance, lighthouse | | backend/algorithmic performance in a specific language (use that language engineer); MCP server development (use mcp-local-docs-engineer) — only web/frontend runtime and load performance | testing | | perses-engineer | Perses observability platform: dashboards, plugins, operator, core development | perses, perses dashboard, perses plugin, perses operator, perses kubernetes, perses CRD, percli, perses migrate, perses dac, PersesDashboard, perses core, contribute perses, perses backend, perses frontend, perses architecture, perses internals, perses project, observability dashboard, perses datasource, perses variable, perses helm, perses k8s, create plugin, panel plugin, datasource plugin, perses plugin development, perses cue schema, plugin schema | | Grafana or Prometheus dashboards, alerting, and PromQL (use prometheus-grafana-engineer); Kubernetes manifests and Helm charts unrelated to Perses (use kubernetes-helm-engineer); TypeScript frontend work outside the Perses codebase (use typescript-frontend-engineer). This agent develops the Perses platform itself: dashboards, plugins, operator, and core. | golang-general-engineer, typescript-frontend-engineer, kubernetes-helm-engineer, prometheus-grafana-engineer | -| php-general-engineer | PHP development: features, debugging, code quality, security, modern PHP 8 | php, laravel, symfony, composer, artisan, eloquent, blade, twig, phpunit, pest, psr-12, psr standards, hybris, sapcc, .php files, doctrine, php-cs-fixer, phpstan, psalm | | running or configuring PHP quality or test tooling in isolation (use php skill); Go SAP Commerce Cloud patterns (use go-patterns skill); Python pip/packaging (use python-general-engineer). This agent writes and debugs PHP features. | workflow, verification-before-completion, systematic-code-review | -| pipeline-orchestrator-engineer | Pipeline orchestration: scaffold multi-component workflows, fan-out/fan-in patterns | create pipeline, new pipeline, scaffold pipeline, build pipeline, pipeline creator | | coordinating agents and dependencies across an in-flight project (use project-coordinator-engineer); authoring one new skill end to end (use skill-creator skill); running an existing workflow or DAG (use workflow skill); routing-table and INDEX consistency (use toolkit-governance-engineer). This agent scaffolds new multi-component pipelines with fan-out/fan-in structure. | workflow, codebase-overview, routing-table-updater | -| pixijs-combat-renderer | PixiJS v8 2D WebGL combat rendering: @pixi/react hybrid canvas, normal maps, GPU particles, post-processing | pixijs, pixi.js, pixi react, @pixi/react, 2D WebGL, GPU particles, combat renderer, normal map 2D, pixi filters, sprite rendering GPU | | CSS or Framer Motion effects that need no WebGL (use combat-effects-upgrade); skeletal character animation and state machines (use rive-skeletal-animator); 3D or Three.js scenes (use threejs-builder skill); general React component architecture (use typescript-frontend-engineer). This agent renders 2D WebGL combat with PixiJS v8. | typescript-frontend-engineer, ui-design-engineer, combat-effects-upgrade | -| project-coordinator-engineer | Multi-agent project coordination: task breakdown, dependency management, progress tracking | coordinate, multi-agent, orchestrate, project, task management, agent coordination | | personal productivity and task-list tools (use business-ops skill, PRODUCTIVITY mode); scaffolding the structure of a new multi-component pipeline (use pipeline-orchestrator-engineer) — this agent coordinates multi-agent project execution and dependency tracking | workflow, subagent-driven-development | -| prometheus-grafana-engineer | Prometheus and Grafana: monitoring, alerting, dashboard design, PromQL optimization | prometheus, grafana, monitoring, alerting, dashboards, metrics, observability | | analyzing business or product metrics data (use data-analysis); Perses dashboards, plugins, operator, or core development (use perses-engineer); deploying the monitoring stack via Kubernetes manifests or Helm charts (use kubernetes-helm-engineer). This agent is Prometheus and Grafana operations, not data analytics. | verification-before-completion, kubernetes-helm-engineer | -| python-general-engineer | Python development: features, debugging, code review, performance | python, .py files, pip, pytest, asyncio, fastapi, django, flask | | PHP development (use php-general-engineer); OpenStack services and Oslo libraries (use python-openstack-engineer); SQLite and Peewee ORM work (use sqlite-peewee-engineer) — this agent handles general Python development | python-quality-gate | -| python-openstack-engineer | OpenStack Python development: Nova, Neutron, Cinder, Oslo libraries, WSGI middleware | openstack, oslo, neutron, nova, cinder, tempest, oslo.config, oslo.messaging | | general Python features, debugging, or web frameworks outside OpenStack (use python-general-engineer); running Python lint, format, and test gates (use python-quality-gate skill); SQLite and Peewee ORM work (use sqlite-peewee-engineer). This agent develops OpenStack services and Oslo libraries. | python-quality-gate, python-general-engineer | -| rabbitmq-messaging-engineer | RabbitMQ: message queue architecture, clustering, high-availability, routing patterns | rabbitmq, messaging, message queue, amqp, event bus | | Kafka, Spark, or stream-processing data pipelines (use data-engineer); application webhook and REST endpoint code (use nodejs-api-engineer); running a broker on Kubernetes via manifests or Helm (use kubernetes-helm-engineer); broker metrics dashboards and alert rules (use prometheus-grafana-engineer). This agent designs RabbitMQ topology, clustering, and high availability. | verification-before-completion | -| react-native-engineer | React Native and Expo development: performance, animations, navigation, native UI patterns | react native, expo, reanimated, flashlist, hermes, react-navigation, gesture handler, native stack | | React web apps (use typescript-frontend-engineer); portfolio and gallery sites for creatives (use react-portfolio-engineer) | universal-quality-gate | +| php-general-engineer | PHP development: features, debugging, code quality, security, modern PHP 8 | programming, laravel, symfony, composer, artisan, eloquent, blade, twig, programmingunit, pest, psr-12, psr standards, hybris, sapcc, .programming files, doctrine, programming-cs-fixer, programmingstan, psalm | | running or configuring PHP quality or test tooling in isolation (use programming skill); Go SAP Commerce Cloud patterns (use programming skill); Python pip/packaging (use python-general-engineer). This agent writes and debugs PHP features. | workflow, testing, review | +| pipeline-orchestrator-engineer | Pipeline orchestration: scaffold multi-component workflows, fan-out/fan-in patterns | create pipeline, new pipeline, scaffold pipeline, build pipeline, pipeline creator | | coordinating agents and dependencies across an in-flight project (use project-coordinator-engineer); authoring one new skill end to end (use toolkit skill); running an existing workflow or DAG (use workflow skill); routing-table and INDEX consistency (use toolkit-governance-engineer). This agent scaffolds new multi-component pipelines with fan-out/fan-in structure. | workflow, assessment, toolkit | +| pixijs-combat-renderer | PixiJS v8 2D WebGL combat rendering: @pixi/react hybrid canvas, normal maps, GPU particles, post-processing | pixijs, pixi.js, pixi react, @pixi/react, 2D WebGL, GPU particles, combat renderer, normal map 2D, pixi filters, sprite rendering GPU | | CSS or Framer Motion effects that need no WebGL (use combat-effects-upgrade); skeletal character animation and state machines (use rive-skeletal-animator); 3D or Three.js scenes (use frontend skill); general React component architecture (use typescript-frontend-engineer). This agent renders 2D WebGL combat with PixiJS v8. | typescript-frontend-engineer, ui-design-engineer, combat-effects-upgrade | +| project-coordinator-engineer | Multi-agent project coordination: task breakdown, dependency management, progress tracking | coordinate, multi-agent, orchestrate, project, task management, agent coordination | | personal productivity and task-list tools (use business-ops skill, PRODUCTIVITY mode); scaffolding the structure of a new multi-component pipeline (use pipeline-orchestrator-engineer) — this agent coordinates multi-agent project execution and dependency tracking | workflow, process | +| prometheus-grafana-engineer | Prometheus and Grafana: monitoring, alerting, dashboard frontend, PromQL optimization | prometheus, grafana, monitoring, alerting, dashboards, metrics, observability | | analyzing business or product metrics data (use data skill); Perses dashboards, plugins, operator, or core development (use perses-engineer); deploying the monitoring stack via Kubernetes manifests or Helm charts (use kubernetes-helm-engineer). This agent is Prometheus and Grafana operations, not data analytics. | testing, kubernetes-helm-engineer | +| python-general-engineer | Python development: features, debugging, code review, performance | python, .py files, pip, pytest, asyncio, fastapi, django, flask | | PHP development (use php-general-engineer); OpenStack services and Oslo libraries (use python-openstack-engineer); SQLite and Peewee ORM work (use sqlite-peewee-engineer) — this agent handles general Python development | code-quality | +| python-openstack-engineer | OpenStack Python development: Nova, Neutron, Cinder, Oslo libraries, WSGI middleware | openstack, oslo, neutron, nova, cinder, tempest, oslo.config, oslo.messaging | | general Python features, debugging, or web frameworks outside OpenStack (use python-general-engineer); running Python lint, format, and test gates (use code-quality skill); SQLite and Peewee ORM work (use sqlite-peewee-engineer). This agent develops OpenStack services and Oslo libraries. | code-quality, python-general-engineer | +| rabbitmq-messaging-engineer | RabbitMQ: message queue architecture, clustering, high-availability, routing patterns | rabbitmq, messaging, message queue, amqp, event bus | | Kafka, Spark, or stream-processing data pipelines (use data-engineer); application webhook and REST endpoint code (use nodejs-api-engineer); running a broker on Kubernetes via manifests or Helm (use kubernetes-helm-engineer); broker metrics dashboards and alert rules (use prometheus-grafana-engineer). This agent frontends RabbitMQ topology, clustering, and high availability. | testing | +| react-native-engineer | React Native and Expo development: performance, animations, navigation, native UI patterns | react native, expo, reanimated, flashlist, hermes, react-navigation, gesture handler, native stack | | React web apps (use typescript-frontend-engineer); portfolio and gallery sites for creatives (use react-portfolio-engineer) | code-quality | | react-portfolio-engineer | React portfolio/gallery sites for creatives: React 18+, Next | portfolio, gallery, react portfolio, art website, image gallery, lightbox | | commerce features such as carts, Stripe, or checkout (use nextjs-ecommerce-engineer); general React or Next.js application architecture (use typescript-frontend-engineer); React Native mobile apps (use react-native-engineer); generating the artwork itself (use image-gen skill). This agent builds portfolio and gallery sites for creatives. | ui-design-engineer, typescript-frontend-engineer | -| research-coordinator-engineer | Research coordination: systematic investigation, multi-source analysis, synthesis | research topic, multi-source analysis, investigate question, synthesize sources, comprehensive analysis, study, examine | | reading/exploring local code (use codebase-overview/read-only-ops); a single delegated research subtask (use research-subagent-executor); running the formal multi-phase research pipeline (use research-pipeline skill) | workflow, subagent-driven-development | -| research-subagent-executor | Research subagent execution: OODA-loop investigation, intelligence gathering, source evaluation | research subtask, delegated research | | decomposing a question or synthesizing across many sources (use research-coordinator-engineer); running the full multi-phase research pipeline (use research-pipeline skill); exploring local code (use codebase-overview skill). This agent executes one delegated research subtask via OODA loops. | research-coordinator-engineer | -| reviewer-code | Code quality review: conventions, naming, dead code, performance, test coverage | code review, review code quality, code conventions, naming review, dead code review, performance review, type design review, test coverage review, config safety review | | Codex second-opinion review or PR lifecycle (use pr-workflow skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent reviews code quality: conventions, naming, dead code, test coverage | workflow, parallel-code-review, systematic-code-review | -| reviewer-domain | Domain-specific review: ADR compliance, business logic, SAP CC structural, pragmatic builder | adr compliance, adr review, architecture decision, decision record, adr check, scope creep, business logic, domain review, requirements, correctness, edge cases, state machine, sapcc structural, go-bits design, sapcc structural review, type export, anti-over-engineering, go-bits usage, builder, production, ops, operational | | naming, dead code, test coverage, and convention review (use reviewer-code); security, concurrency, error handling, observability, and API contracts (use reviewer-system); newcomer, contrarian, pedant, or user-advocate framings (use reviewer-perspectives); full-repo SAP Commerce Cloud compliance audit (use sapcc-audit skill); running a consultation to decide a new architecture decision (use adr-consultation skill). This agent reviews business logic, ADR conformance, and SAP CC structural correctness against domain requirements. | workflow, parallel-code-review, systematic-code-review | -| reviewer-perspectives | Multi-perspective review: newcomer, senior, pedant, contrarian, user advocate, meta-process | newcomer perspective, fresh eyes review, documentation review, production readiness, senior review, skeptical review, technical accuracy, spec compliance, pedantic review, contrarian, alternatives, assumptions, challenge, multi-persona-critique, user impact, user advocate, usability review, is this worth the complexity, user perspective, user experience, meta-process review, system design review, architecture health, single point of failure, indispensable component, complexity audit, authority concentration, reversibility check | | SAP Commerce Cloud full-repo compliance audit (use sapcc-audit skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides multi-perspective review: newcomer, senior, pedant, contrarian views | systematic-code-review, workflow | -| reviewer-system | System-level review: security, concurrency, error handling, observability, API contracts | system review, security review, concurrency review, error handling review, observability review, API contract review, migration safety review, dependency audit, API doc accuracy review | | standalone security scan of git changes (use security-review skill); parallel multi-reviewer code review (use parallel-code-review skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides system-level review: concurrency, error handling, observability, API contracts | workflow, systematic-code-review, parallel-code-review, go-patterns | -| rive-skeletal-animator | Rive skeletal animation: @rive-app/react-canvas, state machines, character pipelines, combat integration | rive, rive-app, skeletal animation, character animation, spine animation, animation state machine, idle animation, hit reaction animation, @rive-app/react-canvas, .riv files | | GPU particle and WebGL combat rendering (use pixijs-combat-renderer); CSS or Framer Motion effects (use combat-effects-upgrade); 2D sprite-sheet generation pipelines (use game-sprite-pipeline skill); 3D scene animation (use threejs-builder skill). This agent integrates Rive skeletal animation, state machines, and character pipelines. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | +| research-coordinator-engineer | Research coordination: systematic investigation, multi-source analysis, synthesis | research topic, multi-source analysis, investigate question, synthesize sources, comprehensive analysis, study, examine | | reading/exploring local code (use assessment skill); a single delegated research subtask (use research-subagent-executor); running the formal multi-phase research pipeline (use research skill) | workflow, process | +| research-subagent-executor | Research subagent execution: OODA-loop investigation, intelligence gathering, source evaluation | research subtask, delegated research | | decomposing a question or synthesizing across many sources (use research-coordinator-engineer); running the full multi-phase research pipeline (use research skill); exploring local code (use assessment skill). This agent executes one delegated research subtask via OODA loops. | research-coordinator-engineer | +| reviewer-code | Code quality review: conventions, naming, dead code, performance, test coverage | code review, review code quality, code conventions, naming review, dead code review, performance review, type frontend review, test coverage review, config safety review | | Codex second-opinion review or PR lifecycle (use pr-workflow skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent reviews code quality: conventions, naming, dead code, test coverage | workflow, review | +| reviewer-domain | Domain-specific review: ADR compliance, business logic, SAP CC structural, pragmatic builder | adr compliance, adr review, architecture decision, decision record, adr check, scope creep, business logic, domain review, requirements, correctness, edge cases, state machine, sapcc structural, go-bits frontend, sapcc structural review, type export, anti-over-engineering, go-bits usage, builder, production, ops, operational | | naming, dead code, test coverage, and convention review (use reviewer-code); security, concurrency, error handling, observability, and API contracts (use reviewer-system); newcomer, contrarian, pedant, or user-advocate framings (use reviewer-perspectives); full-repo SAP Commerce Cloud compliance audit (use domain skill); running a consultation to decide a new architecture decision (use assessment skill). This agent reviews business logic, ADR conformance, and SAP CC structural correctness against domain requirements. | workflow, review | +| reviewer-perspectives | Multi-perspective review: newcomer, senior, pedant, contrarian, user advocate, meta-process | newcomer perspective, fresh eyes review, documentation review, production readiness, senior review, skeptical review, technical accuracy, spec compliance, pedantic review, contrarian, alternatives, assumptions, challenge, assessment, user impact, user advocate, usability review, is this worth the complexity, user perspective, user experience, meta-process review, system frontend review, architecture health, single point of failure, indispensable component, complexity audit, authority concentration, reversibility check | | SAP Commerce Cloud full-repo compliance audit (use domain skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides multi-perspective review: newcomer, senior, pedant, contrarian views | review, workflow | +| reviewer-system | System-level review: security, concurrency, error handling, observability, API contracts | system review, security review, concurrency review, error handling review, observability review, API contract review, migration safety review, dependency audit, API doc accuracy review | | standalone security scan of git changes (use security skill); parallel multi-reviewer code review (use review skill); business-logic correctness, ADR conformance, or SAP CC structural review (use reviewer-domain) — this agent provides system-level review: concurrency, error handling, observability, API contracts | workflow, review, programming | +| rive-skeletal-animator | Rive skeletal animation: @rive-app/react-canvas, state machines, character pipelines, combat integration | rive, rive-app, skeletal animation, character animation, spine animation, animation state machine, idle animation, hit reaction animation, @rive-app/react-canvas, .riv files | | GPU particle and WebGL combat rendering (use pixijs-combat-renderer); CSS or Framer Motion effects (use combat-effects-upgrade); 2D sprite-sheet generation pipelines (use game-dev skill); 3D scene animation (use frontend skill). This agent integrates Rive skeletal animation, state machines, and character pipelines. | typescript-frontend-engineer, ui-design-engineer, pixijs-combat-renderer | | sqlite-peewee-engineer | SQLite with Peewee ORM: model definition, query optimization, migrations, transactions | peewee, sqlite, ORM, python database, playhouse | | server databases like Postgres/MySQL (use database-engineer); general Python features or debugging outside the data layer (use python-general-engineer) | python-general-engineer, database-engineer | -| swift-general-engineer | Swift development: iOS, macOS, server-side Swift, SwiftUI, concurrency, testing | swift, ios, macos, xcode, swiftui, uikit, appkit, watchos, tvos, visionos, vapor, spm, swift-package-manager, swiftlint, swiftformat, xctest, swift actor, swift sendable, swift-combine, swiftdata, coredata | | Swift concurrency patterns or test-only work in isolation (use swift skill) — this agent handles full Swift development | workflow, verification-before-completion, systematic-code-review | -| system-upgrade-engineer | Systematic toolkit upgrades: adapt agents, skills, hooks when Claude Code ships updates | upgrade agents, system upgrade, claude update, upgrade skills, adapt workflow, apply claude update, apply update, system health, update system, new claude version, apply retro | | editing one skill, routing entry, or ADR (use toolkit-governance-engineer); regenerating routing INDEX files (use routing-table-updater skill); writing a new Python hook implementation (use hook-development-engineer); explaining which agent or workflow to run (use workflow-help skill). This agent adapts agents, skills, and hooks across the fleet when Claude Code ships a release. | toolkit-evolution, agent-evaluation, codebase-overview, routing-table-updater, pr-workflow | -| technical-documentation-engineer | Technical documentation: API docs, system architecture, runbooks, enterprise standards | API documentation, technical docs, documentation validation, integration guide | | explainers, opinion pieces, or long-form articles for a public audience (use technical-journalist-writer); detecting drift between existing docs and code (use docs-sync-checker skill); building a local documentation MCP server (use mcp-local-docs-engineer). This agent writes API references, architecture docs, runbooks, and integration guides. | verification-before-completion | -| technical-journalist-writer | Technical journalism: explainers, opinion pieces, analysis articles, long-form content | technical article, technical writing, journalist voice, technical writer, technical journalism | | API references, runbooks, or architecture documentation (use technical-documentation-engineer); drafting through a stored voice profile (use voice-writer skill); building or validating a voice profile (use create-voice or voice-validator skills); SEO, campaign, and launch copy (use marketing skill). This agent writes technical journalism: explainers, opinion, and analysis. | voice-writer | -| testing-automation-engineer | Testing automation: Vitest, Playwright, E2E, coverage enforcement, CI/CD integration | testing, E2E, playwright, vitest, test automation, visual regression | | Playwright-only E2E test authoring (use e2e-testing skill) — this agent covers full test automation strategy including Vitest, coverage, and CI integration | test-driven-development, e2e-testing | -| toolkit-governance-engineer | Toolkit governance: edit skills, update routing tables, manage ADR lifecycle, enforce standards | edit skill, routing table governance, ADR management, toolkit maintenance, check coverage, skill compliance, hook standardization, cross-component consistency, create skill, new skill, scaffold skill, build a skill, create a skill | | authoring one new skill end-to-end (use skill-creator); mechanically regenerating routing INDEX files (use the routing-table-updater skill); adapting the fleet to a new Claude Code release (use system-upgrade-engineer); writing a Python hook implementation (use hook-development-engineer); scaffolding a new multi-component pipeline (use pipeline-orchestrator-engineer). This agent governs fleet-wide policy, routing consistency, and ADR conformance, not single-skill scaffolding or index sync. | adr-consultation, routing-table-updater, docs-sync-checker, skill-creator | -| typescript-debugging-engineer | TypeScript debugging: race conditions, async/await issues, type errors, runtime exceptions | typescript debug, async bug, race condition, type error, production error, memory leak | | building components, state management, or build configuration (use typescript-frontend-engineer); running tsc and clearing reported type errors mechanically (use typescript-check skill); React Native runtime issues (use react-native-engineer); Core Web Vitals and bundle performance (use performance-optimization-engineer). This agent diagnoses TypeScript race conditions, async bugs, type failures, and production runtime exceptions. | workflow, typescript-frontend-engineer | -| typescript-frontend-engineer | TypeScript frontend architecture: type-safe components, state management, build optimization | typescript, react, next.js, frontend, .tsx, .ts, zod | | React Native mobile apps (use react-native-engineer); diagnosing race conditions, async bugs, or production runtime exceptions (use typescript-debugging-engineer); e-commerce carts, Stripe, and checkout flows (use nextjs-ecommerce-engineer); portfolio and gallery sites (use react-portfolio-engineer). This agent builds TypeScript frontend architecture: type-safe components, state, and build configuration. | universal-quality-gate, typescript-check | -| ui-design-engineer | UI/UX design: design systems, responsive layouts, accessibility, animations | UI, design, tailwind, accessibility, responsive, animations, design system | | non-visual design: API, schema, or system design (use the matching engineer); writing design docs (use a planning skill); design methodology, user-research analysis, or accessibility audit documentation (use design skill); combat visual effects and animation juice (use combat-effects-upgrade). This agent is visual UI/UX implementation only. | distinctive-frontend-design, typescript-frontend-engineer | +| swift-general-engineer | Swift development: iOS, macOS, server-side Swift, SwiftUI, concurrency, testing | programming, ios, macos, xcode, programmingui, uikit, appkit, watchos, tvos, visionos, vapor, spm, programming-package-manager, programminglint, programmingformat, xctest, programming actor, programming sendable, programming-combine, programmingdata, coredata | | Swift concurrency patterns or test-only work in isolation (use programming skill) — this agent handles full Swift development | workflow, testing, review | +| system-upgrade-engineer | Systematic toolkit upgrades: adapt agents, skills, hooks when Claude Code ships updates | upgrade agents, system upgrade, claude update, upgrade skills, adapt workflow, apply claude update, apply update, system health, update system, new claude version, apply process | | editing one skill, routing entry, or ADR (use toolkit-governance-engineer); regenerating routing INDEX files (use toolkit skill); writing a new Python hook implementation (use hook-development-engineer); explaining which agent or workflow to run (use workflow-help skill). This agent adapts agents, skills, and hooks across the fleet when Claude Code ships a release. | toolkit, toolkit, assessment, toolkit, pr-workflow | +| technical-documentation-engineer | Technical documentation: API docs, system architecture, runbooks, enterprise standards | API documentation, technical docs, documentation validation, integration guide | | explainers, opinion pieces, or long-form articles for a public audience (use technical-journalist-writer); detecting drift between existing docs and code (use docs-sync-checker skill); building a local documentation MCP server (use mcp-local-docs-engineer). This agent writes API references, architecture docs, runbooks, and integration guides. | testing | +| technical-journalist-writer | Technical journalism: explainers, opinion pieces, analysis articles, long-form content | technical article, technical writing, journalist voice, technical writer, technical journalism | | API references, runbooks, or architecture documentation (use technical-documentation-engineer); drafting through a stored voice profile (use writing skill); building or validating a voice profile (use writing skill); SEO, campaign, and launch copy (use content skill). This agent writes technical journalism: explainers, opinion, and analysis. | writing | +| testing-automation-engineer | Testing automation: Vitest, Playwright, E2E, coverage enforcement, CI/CD integration | testing, E2E, playwright, vitest, test automation, visual regression | | Playwright-only E2E test authoring (use testing skill) — this agent covers full test automation strategy including Vitest, coverage, and CI integration | testing, testing | +| toolkit-governance-engineer | Toolkit governance: edit skills, update routing tables, manage ADR lifecycle, enforce standards | edit skill, routing table governance, ADR management, toolkit maintenance, check coverage, skill compliance, hook standardization, cross-component consistency, create skill, new skill, scaffold skill, build a skill, create a skill | | authoring one new skill end-to-end (use toolkit skill); mechanically regenerating routing INDEX files (use the toolkit skill); adapting the fleet to a new Claude Code release (use system-upgrade-engineer); writing a Python hook implementation (use hook-development-engineer); scaffolding a new multi-component pipeline (use pipeline-orchestrator-engineer). This agent governs fleet-wide policy, routing consistency, and ADR conformance, not single-skill scaffolding or index sync. | assessment, toolkit, docs-sync-checker | +| typescript-debugging-engineer | TypeScript debugging: race conditions, async/await issues, type errors, runtime exceptions | typescript debug, async bug, race condition, type error, production error, memory leak | | building components, state management, or build configuration (use typescript-frontend-engineer); running tsc and clearing reported type errors mechanically (use programming skill); React Native runtime issues (use react-native-engineer); Core Web Vitals and bundle performance (use performance-optimization-engineer). This agent diagnoses TypeScript race conditions, async bugs, type failures, and production runtime exceptions. | workflow, typescript-frontend-engineer | +| typescript-frontend-engineer | TypeScript frontend architecture: type-safe components, state management, build optimization | typescript, react, next.js, frontend, .tsx, .ts, zod | | React Native mobile apps (use react-native-engineer); diagnosing race conditions, async bugs, or production runtime exceptions (use typescript-debugging-engineer); e-commerce carts, Stripe, and checkout flows (use nextjs-ecommerce-engineer); portfolio and gallery sites (use react-portfolio-engineer). This agent builds TypeScript frontend architecture: type-safe components, state, and build configuration. | code-quality, programming | +| ui-design-engineer | UI/UX frontend: frontend systems, responsive layouts, accessibility, animations | UI, frontend, tailwind, accessibility, responsive, animations, frontend system | | non-visual frontend: API, schema, or system frontend (use the matching engineer); writing frontend docs (use a workflow skill); frontend methodology, user-research analysis, or accessibility audit documentation (use frontend skill); combat visual effects and animation juice (use combat-effects-upgrade). This agent is visual UI/UX implementation only. | frontend, typescript-frontend-engineer | -## SKILLS (43) +## SKILLS (44) | name | when to call (description) | triggers | FORCE | not_for | pairs_with | |---|---|---|---|---|---| +| accounts | Manage multiple Claude Code accounts: add, list, check, launch, and install shell aliases for 10+ isolated CLAUDE_CONFIG_DIR profiles. | multiple accounts, multiple claude accounts, CLAUDE_CONFIG_DIR, claude account, switch accounts, account alias, claude login, c1 c2 aliases, claude max account, manage accounts, add claude account | | API key rotation (use secrets management), Anthropic billing (use the web console) | deploy | | architecture-deepening | Improve architecture across modules by deepening interfaces. | improve architecture, improve codebase architecture, improve the codebase architecture, find architecture improvements, deepen architecture, find shallow modules, architecture improvement, module depth analysis, deepening opportunities, improve module interfaces, architecture deepening | | local cleanup/refactoring (workflow or planning), feature design (feature-lifecycle), architecture overview/explanation (codebase-overview), or vague complexity reduction; requires cross-module interface or caller-burden evidence | review, assessment | | assessment | Assessment: read-only inspection, codebase overview, value analysis, health checks, ADR consultation, decision analysis, multi-perspective critique. | inspect without changing, read-only, audit current state, onboard to codebase, codebase structure, give me an overview, summarize this repo, repo value analysis, what can we learn from, service status, check health, is service running, validate endpoints, consult on ADR, architecture consultation, adr consultation, help me decide, decision matrix, pros and cons, trade-offs, critique these ideas, devil's advocate, stress test proposals, roast this, poke holes in this | | code review with findings (use review), building or fixing (use workflow) | review, workflow, security | | auto-dream | Background memory consolidation — overnight review, merge, and injection payload for memory files. | dream, consolidate memories, clean up memories, memory maintenance, memory consolidation, deduplicate memories | | | | @@ -77,13 +77,13 @@ CI check: `python3 scripts/generate-routing-map.py --check`. | do | Classify user requests and route to the correct agent + skill. | route task, classify request, which agent, delegate to skill, smart router | | | | | docs-sync-checker | Detect documentation drift against filesystem state. | check doc drift, sync documentation, stale docs, documentation drift, README outdated | | | toolkit, assessment | | domain | Domain-specific: SAP Commerce, OpenSearch detection, WordPress validation, enterprise search. | sapcc review, sapcc audit, sapcc compliance, sapcc standards, siem detection, sigma rule, mitre att&ck mapping, detection engineering, opensearch detection, anomaly detection rule, soc escalation, validate wordpress post, check live post, wordpress post validation, enterprise search, search relevance, search ranking, BM25, query understanding, search quality, OpenSearch, Elasticsearch, vector search, hybrid search, search tuning | | general code review (use review), general security (use security) | golang-general-engineer, opensearch-elasticsearch-engineer, programming | -| frontend | Frontend: UI design, distinctive visual styles, HTML artifacts, Three.js 3D. | design, UX copy, design system, accessibility, WCAG, design handoff, frontend design, visual identity, HTML artifact, make HTML, self-contained HTML, threejs, three.js, 3D web, 3D scene, WebGL, WebGPU, react three fiber, r3f, text animation, card effects, pptx, powerpoint, slide deck, design critique | | game UI (use game-dev), backend work (use workflow), brand or logo design without code | typescript-frontend-engineer, ui-frontend-engineer, game-dev | +| frontend | Frontend: UI design, distinctive visual styles, HTML artifacts, Three.js 3D. | design, UX copy, design system, accessibility, WCAG, design handoff, frontend design, visual identity, HTML artifact, make HTML, self-contained HTML, threejs, three.js, 3D web, 3D scene, WebGL, WebGPU, react three fiber, r3f, text animation, card effects, pptx, powerpoint, slide deck, design critique | | game UI (use game-dev), backend work (use workflow), brand or logo design without code | typescript-frontend-engineer, ui-design-engineer, game-dev | | game-dev | Game development: design, asset pipelines, Phaser, vanilla JS frontend, GM mode, match booking. | game design, game design audit, game improvement, core loop, game feel, game balance, game economy, make game, game pipeline, game audio, deploy game, phaser, 2d game, platformer, arcade physics, tilemap, AI sprite, generate sprite, spritesheet pipeline, animated spritesheet, large GM implementation, GM systems overhaul, pixel art, screen shake, game qa | yes | non-game frontend (use frontend), non-game backend (use workflow) | frontend, workflow | | github | GitHub: notification triage, profile rule extraction. | github notifications, triage notifications, check notifications, notification cleanup, github inbox, triage issues, triage pull requests, pr queue, github profile rules, profile conventions | | git commit/push/PR operations (use pr-workflow) | | | grill-jev | Broad Jev interrogation: generate up to 50 context-specific questions about a plan, spec, design, code artifact, or any topic — using all primitive... | grill, validate plan, validate spec, validate design, interrogate, stress test plan, stress test spec, 50 questions, deep validation, grill this, grill the plan, jev interrogation, sanity check plan, sanity check spec, check my plan, check my design, plan validation, spec validation, design validation, architecture review jev, verify plan, verify spec, challenge plan, challenge design, find holes, find gaps in plan | yes | Writing new Jev programs (use building-with-jev). Routing requests (use do). Code review for quality/style (use review). Security scanning (use security). This skill runs a broad interrogation battery against a supplied artifact — it does not build Jev programs. | building-with-jev, review | | image-gen | AI image generation: Gemini and Nano Banana backends; single/series/batch workflows with prompt-to-disk. | generate image, create image, image generation, AI image, gemini image, make image, draw, illustrate, sprite generation, card art, batch generation, series of images, character art, pixel art, image post-processing, make a picture, create art, generate a picture, make art, generate artwork, create artwork | | HTML visualization or charts (use html-artifact), or deterministic non-AI palette/matrix pixel art (use game-asset-generator) | python-general-engineer, game-dev | | integration-checker | Verify cross-component wiring and data flow. | integration check, check integration, verify wiring, are components connected, check connections, integration-checker, wiring check | | | workflow, review | -| jev-design | Plan React interfaces with shadcn components and style recipes selected from bounded catalogs by live Jev. | jev design, design with jev, shadcn component plan, choose shadcn components, shadcn ui design, component style recipe | | | frontend, building-with-jev | +| jev-design | Design React interfaces with the full shadcn/ui catalog. | jev design, design with jev, shadcn component plan, choose shadcn components, shadcn ui design, component style recipe, which shadcn component, shadcn theme recipe, pick shadcn components, design a shadcn screen | | | frontend, building-with-jev | | joy-check | Validate content framing on joy-grievance spectrum. | joy check, check framing, tone check, negative framing, joy validation, too negative, reframe positively, positive framing check, instruction framing | | | writing, toolkit | | kubernetes | Kubernetes operations: debugging, security, RBAC, and infrastructure tooling. | kubernetes debug, pod failure, pod crashloop, kubectl logs, OOMKilled, pod pending, kubernetes security, k8s RBAC, RBAC setup, pod security policy, network policy, cobalt core, cobaltcore, kvm-exporter, kvm exporter, hypervisor metrics, libvirt exporter, cloud hypervisor | | | assessment, programming, prometheus-grafana-engineer | | markdown-converter | Convert PDF, Office, HTML, data, media, ZIP to Markdown. | convert to markdown, markitdown, extract text from PDF, PDF to markdown, docx to markdown, ingest document, read this PDF, read this document, extract text from document, convert PDF, convert document, pptx to markdown, xlsx to markdown | | | research, domain | @@ -93,14 +93,14 @@ CI check: `python3 scripts/generate-routing-map.py --check`. | programming | Language-specific patterns and tooling: Go, Kotlin, PHP, Swift, TypeScript. | .go, go test, *_test.go, goroutine, sync.Mutex, context.Context, Go code review, Go lint, sapcc, make check, kotlin coroutines, kotlin Flow, suspend function, kotlin testing, kotest, php quality, php code review, PSR standards, phpstan, phpunit, swift concurrency, swift async await, Swift Actor, XCTest, TypeScript check, tsc noEmit, tsc errors | yes | Python (agents handle directly), general debugging (use debugging). When a dedicated language skill exists (php, swift, kotlin), prefer it over this umbrella | golang-general-engineer, kotlin-general-engineer, php-general-engineer, swift-general-engineer, typescript-frontend-engineer | | quick | Tracked lightweight execution with composable rigor flags: --trivial, --discuss, --research, --full. | quick task, small change, ad hoc task, add a flag, small refactor, targeted fix, quick fix, typo fix, fix typo, fix the typo, one-line change, trivial fix, rename variable, rename this variable, update value, fix import, small mistake, small mistake in, mistake in spelling, spelling mistake, spelling fix, fix the spelling, typo in, small fix in, small fix, tiny fix | yes | 'quick' as speed preference, general bug diagnosis requiring investigation | | | research | Research: structured investigation, fact-checking, explanation traces. | research-pipeline, research, formal research, research with artifacts, systematic investigation, research report, gather evidence, fact check, fact-check, verify claims, check facts, verify this quote, is this accurate, is this true, check this claim, verify this, are these numbers right, why did you, explain routing, show trace, decision log, why that agent, explain decision, show decisions, trace log | yes | code review (use review), security audit (use security) | review, writing | -| review | Code review: systematic single-file, parallel multi-reviewer, full-repo audit, PR diff review. | review code, code review, code review methodology, structured review, code audit, review methodology, comprehensive review, parallel review, 3-reviewer review, multi-reviewer, concurrent review, full repo review, review entire repo, codebase health check, review all files, full codebase review, audit the codebase, codebase audit, review whole repo, sweep all source files, repo health, review this PR, review my PR, PR review, diff review | yes | security scanning (use security), linting (use code-quality), git push/commit/PR (use pr-workflow) | security, testing | -| security | Security: review git changes for vulnerabilities, threat-model a system's attack surface, audit supply-chain risks. | security review, review my changes for security, review my changes, review for security, security scan, review for vulnerabilities, scan for vulnerabilities, check for security issues, security issues, audit for vulnerabilities, audit auth, audit vulnerabilities, threat model, security audit, supply chain scan, deny list, security posture, injection scan, surface scan, audit hooks, audit skills | yes | general code review (use review), non-security quality checks (use code-quality) — only for security-specific work: vulnerability scanning, threat modeling, supply-chain audits, auth reviews | review, reviewer-system | -| testing | Testing: TDD, E2E, preferred patterns, verification, agent testing. | TDD, test first, red green refactor, write tests first, test-driven, tests before code, flaky test, brittle test, test smell, test quality issue, slow tests, over-mocking, test agents, agent testing, subagent testing, run vitest, JavaScript tests, TypeScript tests, playwright, E2E test, end-to-end, browser test, verify completion, run tests, final verification | | code review (use review), linting (use code-quality) | review, code-quality, workflow | -| toolkit | Toolkit management: create and evaluate skills and agents, manage routing tables, generate Claude.md. | create skill, create agent, scaffold skill, scaffold agent, new skill, new agent, skill template, agent template, eval skill, evaluate agent, benchmark skill, benchmark agents, compare agents, A/B test agents, optimize description, evolve toolkit, toolkit evolution, self-improve, update routing tables, routing maintenance, generate claude.md, create claude.md, compose skills, bake-off, skill quality | yes | application code (use workflow), code review (use review) | review, workflow | +| review | Code review: systematic single-file, parallel multi-reviewer, full-repo audit, PR diff review. | review code, code review, code review methodology, structured review, code audit, review methodology, comprehensive review, parallel review, 3-reviewer review, multi-reviewer, concurrent review, full repo review, review entire repo, codebase health check, review all files, full codebase review, audit the codebase, codebase audit, review whole repo, sweep all source files, repo health, review this PR, review my PR, review my changes, PR review, diff review | yes | security scanning (use security), linting (use code-quality), git push/commit/PR (use pr-workflow) | security, testing | +| security | Security: review git changes for vulnerabilities, threat-model a system's attack surface, audit supply-chain risks. | security review, review my changes for security, review for security, security scan, review for vulnerabilities, scan for vulnerabilities, check for security issues, security issues, audit for vulnerabilities, audit auth, audit vulnerabilities, threat model, security audit, supply chain scan, deny list, security posture, injection scan, surface scan, audit hooks, audit skills | yes | general code review (use review), non-security quality checks (use code-quality) — only for security-specific work: vulnerability scanning, threat modeling, supply-chain audits, auth reviews | review, reviewer-system | +| testing | Testing: TDD, E2E, preferred patterns, test-value audits, verification, agent testing. | TDD, test first, red green refactor, write tests first, test-driven, tests before code, flaky test, brittle test, test smell, test quality issue, slow tests, over-mocking, test audit, audit tests, low-value tests, prune tests, test bloat, delete useless tests, test agents, agent testing, subagent testing, run vitest, JavaScript tests, TypeScript tests, playwright, E2E test, end-to-end, browser test, verify completion, run tests, final verification | | code review (use review), linting (use code-quality) | review, code-quality, workflow | +| toolkit | Toolkit management: create and evaluate skills and agents, manage routing tables, generate Claude.md. | create skill, create agent, scaffold skill, scaffold agent, new skill, new agent, skill template, agent template, eval skill, evaluate agent, benchmark skill, benchmark agents, compare agents, A/B test agents, optimize description, evolve toolkit, toolkit evolution, self-improve, update routing tables, routing maintenance, generate claude.md, create claude.md, compose skills, bake-off, skill quality, weaker model, weak model uplift, improve skill for weaker models | yes | application code (use workflow), code review (use review) | review, workflow | | video-editing | Video editing pipeline: cut footage, assemble clips via FFmpeg and Remotion. | edit video, cut footage, make vlog, screen recording, video workflow, ffmpeg, remotion, demo video, make a clip, assemble clips, video editing, image to video, audio visualization, static video, mp4 from image, music video, podcast video, video from image, combine image audio, album art video, cover art video, video transcript, youtube transcript, extract transcript, download subtitles, what does this video say | | | typescript-frontend-engineer, research | | workflow | Structured work: multi-phase tasks, feature builds, planning, objective loops, hill climbing. | workflow, multi-phase task, feature design, feature plan, feature implement, build feature end to end, full feature lifecycle, write spec, define requirements, create plan, create tasks, keep working until, iterate until done, drive this to done, make this faster, speed this up, reduce latency, profile and optimize, hill climb on this metric, tidy up, clean up, untangle, reorganize, structured pipeline, phased execution | yes | code review (use review), testing (use testing), security (use security) | review, testing, security, pr-workflow | | workflow-help | Interactive guide to workflow system: agents, skills, routing, execution patterns. | how does routing work, what skills exist, system help, explain workflow, I am stuck, toolkit help | yes | | workflow, do | -| writing | Writing: voice creation and validation, prose editing, anti-AI cleanup, professional communication, translation. | write article, blog post, write in voice, voice pipeline, voice-writer, voice writer, validate voice, voice fidelity, write email, draft memo, executive summary, status update, meeting notes, pushback email, disagree professionally, translate, translation, localize, draft article, generate content, professional communication, draft response | yes | code comments (use code-quality), API documentation (use docs-sync-checker), scheduling or calendaring content (use content), code review (use review) | joy-check, content | +| writing | Writing: voice creation and validation, prose editing, anti-AI cleanup, professional communication, translation. | write article, blog post, write in voice, voice pipeline, voice writer, validate voice, voice fidelity, write email, draft memo, executive summary, status update, meeting notes, pushback email, disagree professionally, translate, translation, localize, draft article, generate content, professional communication, draft response | yes | code comments (use code-quality), API documentation (use docs-sync-checker), scheduling or calendaring content (use content), code review (use review) | joy-check, content | ## PIPELINES (25) From 94b9fde035d10208f98d1a0ce9545d21bcf343c8 Mon Sep 17 00:00:00 2001 From: thecodingshrimp Date: Wed, 30 Sep 2026 18:20:10 +0200 Subject: [PATCH 6/6] remove dead depth-3 hybrid-dir else branch, document validator/indexer depth asymmetry, add overwrite-warning test --- scripts/generate-skill-index.py | 11 --------- scripts/tests/test_generate_skill_index.py | 28 ++++++++++++++++++++++ scripts/validate-skill-names.py | 2 ++ 3 files changed, 30 insertions(+), 11 deletions(-) diff --git a/scripts/generate-skill-index.py b/scripts/generate-skill-index.py index 594d74497..88f0fdca2 100755 --- a/scripts/generate-skill-index.py +++ b/scripts/generate-skill-index.py @@ -455,17 +455,6 @@ def _process_skill_dir(skill_dir: Path, skill_file_override: Path | None = None) _process_skill_dir(nested) elif (nested / "skill" / "SKILL.md").exists(): _process_skill_dir(nested, skill_file_override=nested / "skill" / "SKILL.md") - else: - # Third level within hybrid dir: e.g., skills/process/subcat/{name}/SKILL.md - for deep in sorted(nested.iterdir()): - if not deep.is_dir(): - continue - if deep.is_symlink() and not include_private: - continue - if (deep / "SKILL.md").exists(): - _process_skill_dir(deep) - elif (deep / "skill" / "SKILL.md").exists(): - _process_skill_dir(deep, skill_file_override=deep / "skill" / "SKILL.md") # Check for nested skill/SKILL.md layout (e.g., voice-example/skill/SKILL.md) elif (child / "skill" / "SKILL.md").exists(): _process_skill_dir(child, skill_file_override=child / "skill" / "SKILL.md") diff --git a/scripts/tests/test_generate_skill_index.py b/scripts/tests/test_generate_skill_index.py index 224bc40a9..8d2e69344 100755 --- a/scripts/tests/test_generate_skill_index.py +++ b/scripts/tests/test_generate_skill_index.py @@ -401,3 +401,31 @@ def test_hybrid_dir_top_and_nested_both_indexed(self, tmp_path: Path) -> None: assert "process" in index["skills"], "Top-level hybrid skill must be indexed" assert "pr-workflow" in index["skills"], "Nested skill in hybrid dir must be indexed" + + +class TestOverwriteWarning: + """A duplicate skill name emits an overwrite warning.""" + + def test_duplicate_name_emits_warning(self, tmp_path: Path) -> None: + """Two skill dirs with the same name: in frontmatter triggers the overwrite-warning branch.""" + skills_dir = tmp_path / "skills" + # Two different dir names, same name: in frontmatter so both map to "shared-name". + dir_a = skills_dir / "cat-a" / "shared-skill" + dir_a.mkdir(parents=True) + (dir_a / "SKILL.md").write_text(_SKILL_FRONTMATTER.format(name="shared-name")) + + dir_b = skills_dir / "cat-b" / "shared-skill" + dir_b.mkdir(parents=True) + (dir_b / "SKILL.md").write_text(_SKILL_FRONTMATTER.format(name="shared-name")) + + index, warnings = gsi.generate_index( + source_dir=skills_dir, + dir_prefix="skills", + collection_key="skills", + repo_root=tmp_path, + ) + + assert "shared-name" in index["skills"], "The skill must still be indexed" + assert any("shared-name" in w for w in warnings), ( + f"Expected an overwrite warning for 'shared-name'; got warnings: {warnings}" + ) diff --git a/scripts/validate-skill-names.py b/scripts/validate-skill-names.py index 9e81f81b4..fd3643c08 100644 --- a/scripts/validate-skill-names.py +++ b/scripts/validate-skill-names.py @@ -39,6 +39,8 @@ def find_duplicates(repo: Path) -> dict[str, list[str]]: if (top / "SKILL.md").is_file(): seen.setdefault(top.name, []).append(top.relative_to(repo).as_posix()) # Hybrid dir: also scan immediate subdirs for nested skills. + # Intentionally scans only 1 level deep (vs. generate-skill-index.py's 2 levels): + # this is a fast-fail gate, not an exhaustive mirror of the indexer. for child in sorted(top.iterdir()): if child.is_dir() and (child / "SKILL.md").is_file(): seen.setdefault(child.name, []).append(child.relative_to(repo).as_posix())