Framework-agnostic tool primitives"] - agents["llm-coding-tools-agents
Agent markdown loader + runtime"] - serdesai["llm-coding-tools-serdesai
SerdesAI framework integration"] - bubblewrap["llm-coding-tools-bubblewrap
Linux sandbox profiles"] - modelsdev["llm-coding-tools-models-dev
models.dev catalog sync"] + core["reloaded-code-core
Framework-agnostic tool primitives"] + agents["reloaded-code-agents
Agent markdown loader + runtime"] + serdesai["reloaded-code-serdesai
SerdesAI framework integration"] + bubblewrap["reloaded-code-bubblewrap
Linux sandbox profiles"] + modelsdev["reloaded-code-models-dev
models.dev catalog sync"] agents --> core serdesai --> core @@ -24,7 +24,7 @@ graph TD ## Layer overview -### llm-coding-tools-core +### reloaded-code-core The foundation. Contains every tool implementation as a plain function (`read_file`, `write_file`, `edit_file`, etc.), plus supporting types: @@ -39,7 +39,7 @@ Core is **framework-agnostic**: it has no dependencies on any specific LLM framework. Your integration layer wraps these functions into framework-specific tool types. -### llm-coding-tools-agents +### reloaded-code-agents Loads agent definitions from markdown files with YAML frontmatter. Provides: @@ -52,7 +52,7 @@ files are drop-in compatible, but [not identical](migration.md). The most notable difference is **default-deny** permissions: tools must be explicitly allowed. -### llm-coding-tools-serdesai +### reloaded-code-serdesai The ready-to-use integration layer for the [SerdesAI] framework. It: @@ -64,7 +64,7 @@ The ready-to-use integration layer for the [SerdesAI] framework. It: If you use [SerdesAI], this is the only crate you need. If you use a different framework, build your own adapters using core. -### llm-coding-tools-bubblewrap +### reloaded-code-bubblewrap Linux-only. Builds and manages [bubblewrap] sandbox profiles for shell command isolation. Two presets: @@ -72,7 +72,7 @@ isolation. Two presets: - **Public Bot** - for untrusted input (no network, minimal filesystem) - **Trusted Maintenance** - for trusted automation (read-only host `/`, network on) -### llm-coding-tools-models-dev +### reloaded-code-models-dev Syncs the online [models.dev](https://models.dev) catalog into a compact `ModelCatalog`. Features: @@ -86,10 +86,10 @@ Syncs the online [models.dev](https://models.dev) catalog into a compact There are two integration paths: -**Path A: Use [SerdesAI]** - Add `llm-coding-tools-serdesai`, attach tools to an +**Path A: Use [SerdesAI]** - Add `reloaded-code-serdesai`, attach tools to an `AgentBuilder`, and run. This is the fastest path to a working agent. -**Path B: Bring your own framework** - Depend on `llm-coding-tools-core`, +**Path B: Bring your own framework** - Depend on `reloaded-code-core`, implement your framework's tool trait by calling the core functions, and use `SystemPromptBuilder` to generate the system prompt. See [Custom Framework](guides/custom-framework.md) for a walkthrough. diff --git a/docs/src/comparison.md b/docs/src/comparison.md index 0fe877d0..0c14c75b 100644 --- a/docs/src/comparison.md +++ b/docs/src/comparison.md @@ -1,8 +1,8 @@ # Comparison with OpenCode -This page breaks down where llm-coding-tools overlaps with [OpenCode], where it +This page breaks down where reloaded-code overlaps with [OpenCode], where it diverges, and which to pick. [OpenCode] excels at interactive development; -llm-coding-tools is for embedding agent tools into your own applications. +reloaded-code is for embedding agent tools into your own applications. ## At a glance @@ -12,7 +12,7 @@ llm-coding-tools is for embedding agent tools into your own applications.
-
llm-coding-tools
+ReloadedCode
Production-grade coding agent tools in Rust.
~10 MiB. No TUI. Embed it anywhere.
@@ -14,17 +14,17 @@ hide:
- 
- 
- 
- 
+ 
+ 
+ 
+ 
@@ -32,7 +32,7 @@ hide:
## Why this project?
-llm-coding-tools started as "an OpenCode for servers." Headless,
+reloaded-code started as "an OpenCode for servers." Headless,
sandboxed, and cheap to host for non-commercial use.
[OpenCode] is a great interactive coding agent, but it's a **TypeScript
@@ -40,7 +40,7 @@ application** that uses
@@ -113,10 +113,10 @@ Shell sandboxing. Default-deny permissions. ~10 MiB footprint.
```toml
[dependencies]
-llm-coding-tools-serdesai = "0.2"
-llm-coding-tools-agents = "0.1"
-llm-coding-tools-core = "0.2"
-llm-coding-tools-models-dev = "0.1"
+reloaded-code-serdesai = "0.2"
+reloaded-code-agents = "0.1"
+reloaded-code-core = "0.2"
+reloaded-code-models-dev = "0.1"
tokio = { version = "1", features = ["full"] }
```
@@ -144,10 +144,10 @@ You are a coding assistant. Use the available tools to complete the user's task.
**3.** Load the catalog, build the agent, and run:
```rust
-use llm_coding_tools_agents::{AgentCatalog, AgentLoader, AgentRuntimeBuilder};
-use llm_coding_tools_core::CredentialResolver;
-use llm_coding_tools_models_dev::ModelsDevCatalog;
-use llm_coding_tools_serdesai::{AgentBuildContext, AgentDefaults};
+use reloaded_code_agents::{AgentCatalog, AgentLoader, AgentRuntimeBuilder};
+use reloaded_code_core::CredentialResolver;
+use reloaded_code_models_dev::ModelsDevCatalog;
+use reloaded_code_serdesai::{AgentBuildContext, AgentDefaults};
use std::{path::PathBuf, sync::Arc};
#[tokio::main]
@@ -184,23 +184,23 @@ dependency setup and an alternate path without agent files.
-
-
-
+
+
+
+
-
core
+core
Framework-agnostic tools for building coding agents. File operations, search, shell, permissions, system prompts - use with any LLM framework.
-
agents
+agents
Load agent markdown files based on OpenCode's schema into a typed catalog. Default-deny permissions with granular path matching.
-
serdesai
+serdesai
Ready-to-use SerdesAI (LLM serialization framework) integration. 15 LLM provider adapters, multi-agent task delegation with recursion depth limits.
-
bubblewrap
+bubblewrap
Sandbox shell execution on Linux. Network-isolated, filesystem-filtered profiles for untrusted input. Two presets included.
-
models-dev
+models-dev
Sync the models.dev catalog. ETag caching, offline fallback. ~3000 models in ~24 KiB.