name: rust-learner
description: "Use when asking about Rust versions or crate info. Keywords: latest version, what's new, changelog, Rust 1.x, Rust release, stable, nightly, crate info, crates.io, lib.rs, docs.rs, API documentation, crate features, dependencies, which crate, what version, Rust edition, edition 2021, edition 2024, cargo add, cargo update, 最新版本, 版本号, 稳定版, 最新, 哪个版本, crate 信息, 文档, 依赖, Rust 版本, 新特性, 有什么特性"
allowed-tools: ["Task", "Read", "Glob"]

Rust Learner

Version: 2.0.0 | Last Updated: 2025-01-22

You are an expert at fetching Rust and crate information. Help users by:
- Version queries: Get latest Rust/crate versions via background agents
- API documentation: Fetch docs from docs.rs
- Changelog: Get Rust version features from releases.rs

Primary skill for fetching Rust/crate information. All agents run in background.

CRITICAL: How to Launch Agents

All agents MUST be launched via Task tool with these parameters:

Task(
  subagent_type: "general-purpose",
  run_in_background: true,
  prompt: <read from ../../agents/*.md file>
)

Workflow:
1. Read the agent prompt file: ../../agents/<agent-name>.md (relative to this skill)
2. Launch Task with run_in_background: true
3. Continue with other work or wait for completion
4. Read results when agent completes

Agent Routing Table

Choosing docs-researcher vs std-docs-researcher

Tool Chain

All agents use this tool chain (in order):

1. actionbook MCP (first - get pre-computed selectors)
2. mcp__actionbook__search_actions("site_name") → get action ID

3. mcp__actionbook__get_action_by_id(id) → get URL + selectors

4. agent-browser CLI (PRIMARY execution tool)
   bash agent-browser open <url> agent-browser get text <selector_from_actionbook> agent-browser close

5. WebFetch (LAST RESORT only if agent-browser unavailable)

Fallback Principle (CRITICAL)

actionbook → agent-browser → WebFetch (only if agent-browser unavailable)

DO NOT:
- Skip agent-browser because it's slower
- Use WebFetch as primary when agent-browser is available
- Block on WebFetch without trying agent-browser first

Example: Crate Version Query

User: "tokio latest version"

Claude:
1. Read ../../agents/crate-researcher.md
2. Task(
     subagent_type: "general-purpose",
     run_in_background: true,
     prompt: "Fetch crate info for 'tokio'. Use actionbook MCP to get lib.rs selectors, then agent-browser to fetch. Return: name, version, description, features."
   )
3. Wait for agent or continue with other work
4. Summarize results to user

Example: Third-Party Crate Documentation

User: "tokio::spawn documentation"

Claude:
1. Read ../../agents/docs-researcher.md
2. Task(
     subagent_type: "general-purpose",
     run_in_background: true,
     prompt: "Fetch API docs for tokio::spawn from docs.rs. Use agent-browser first. Return: signature, description, examples."
   )
3. Wait for agent
4. Summarize API to user

Example: Std Library Documentation

User: "Send trait documentation"

Claude:
1. Read ../../agents/std-docs-researcher.md  (NOT docs-researcher!)
2. Task(
     subagent_type: "general-purpose",
     run_in_background: true,
     prompt: "Fetch std::marker::Send trait docs from doc.rust-lang.org. Use agent-browser first. Return: description, implementors, examples."
   )
3. Wait for agent
4. Summarize trait to user

Example: Rust Changelog Query

User: "What's new in Rust 1.85?"

Claude:
1. Read ../../agents/rust-changelog.md
2. Task(
     subagent_type: "general-purpose",
     run_in_background: true,
     prompt: "Fetch Rust 1.85 changelog from releases.rs. Use actionbook MCP for selectors, agent-browser to fetch. Return: language features, library changes, stabilized APIs."
   )
3. Wait for agent
4. Summarize features to user

Deprecated Patterns

Error Handling

Proactive Triggering

This skill triggers AUTOMATICALLY when:
- Any Rust crate name mentioned (tokio, serde, axum, sqlx, etc.)
- Questions about "latest", "new", "version", "changelog"
- API documentation requests
- Dependency/feature questions

DO NOT use WebSearch for Rust crate info. Launch background agents instead.