claude-craft

# Claude Code Extensibility Create and configure the six Claude Code extension points: agents, commands, hooks, skills, rules, and configuration. ## Component Map | Component | Purpose | Location | Invocation | | ------------ | --------------------- | ------------------------------------- | --------------------------- | | **Agents** | Specialized subagents | `agents/*.md` | Task tool (`subagent_type`) | | **Commands** | Reusable prompts | `commands/*.md` | User: `/command-name` | | **Hooks** | Event handlers | `settings.json` or `hooks/hooks.json` | Automatic on events | | **Skills** | Capability packages | `skills/*/SKILL.md` | Auto-triggered by context | | **Rules** | Convention files | `.claude/rules/*.md` | Loaded on demand | | **Config** | Settings & MCP | `settings.json`, `.mcp.json` | Loaded at startup | ### Scope Hierarchy | Scope | Location | Visibility | | -------- | ------------------ | ------------ | | Personal | `~/.claude/` | You only | | Project | `.claude/` or root | Team via git | | Plugin | `<plugin>/` | Plugin users | ## Routing | Task | Use | | ----------------------------------------------- | --------------------------------------- | | Creating agents, commands, hooks, rules, config | This skill | | Creating skills (Claude-specific features) | This skill + `skillcraft` for base spec | | Creating skills (cross-platform spec) | `skillcraft` | | Packaging/distributing plugins | `claude-plugins` | --- ## Agents Specialized subagents with focused expertise, invoked via the Task tool. ### Frontmatter ```yaml --- name: security-reviewer # Required: kebab-case, matches filename description: | # Required: triggers + examples Use this agent for security vulnerability detection. Triggers on security audits, OWASP, injection, XSS. <example> Context: User wants security review. user: "Review auth code for vulnerabilities" assistant: "I'll use the security-reviewer agent." </example> model: inherit # Optional: inherit(default)|haiku|sonnet|opus tools: Glob, Grep, Read # Optional: comma-separated (default: inherit all) disallowedTools: Write, Edit # Optional: deny specific tools skills: tdd-fieldguide, debugging # Optional: auto-load skills (NOT inherited) maxTurns: 50 # Optional: max agentic turns memory: user # Optional: user|project|local — persistent memory hooks: # Optional: lifecycle hooks scoped to this agent PreToolUse: - matcher: "Bash" hooks: - type: command command: "./scripts/validate.sh" --- ``` See [agents/frontmatter.md](references/agents/frontmatter.md) for complete schema. ### Archetypes | Type | Purpose | Typical Tools | | --------------- | ------------------------- | -------------------------------------------------------------------------- | | **Analyzer** | Examine without modifying | `Glob, Grep, Read, Skill, Task, TaskCreate, TaskUpdate, TaskList, TaskGet` | | **Implementer** | Build and modify code | Full access (inherit) | | **Reviewer** | Provide feedback | Read-only + Skill + Task tools | | **Tester** | Create and manage tests | Read, Write, Edit, Bash, ... | | **Researcher** | Find and synthesize info | ..., WebSearch, WebFetch | See [agents/types.md](references/agents/types.md) for details. ### Model Selection | Model | When to Use | | --------- | ---------------------------------------------- | | `inherit` | Recommended default — adapts to parent context | | `haiku` | Fast exploration, simple tasks, low-latency | | `sonnet` | Balanced cost/capability (default if omitted) | | `opus` | Nuanced judgment, security/architecture review | ### Tool Configuration **Baseline** (always include when restricting): ```yaml tools: Glob, Grep, Read, Skill, Task, TaskCreate, TaskUpdate, TaskList, TaskGet ``` Only restrict tools when there's a specific safety reason. See [agents/tools.md](references/agents/tools.md). ### Agent Body Structure ```markdown # Agent Name You are a [role] specializing in [expertise]. ## Process ### Step 1: [Stage] - Action items ## Output Format Structured output spec. ## Constraints **Always:** Required behaviors **Never:** Prohibited actions ``` ### Validation Checklist - [ ] `name` matches filename (kebab-case, 1-3 words) - [ ] `description` has WHAT, WHEN, triggers, 2-3 examples - [ ] `tools` uses comma-separated valid names (case-sensitive) - [ ] Body has clear role, process, output format, constraints ### References | Reference | Content | | --------------------------------------------------------------------- | ------------------------------------ | | [agents/frontmatter.md](references/agents/frontmatter.md) | YAML schema and fields | | [agents/types.md](references/agents/types.md) | Archetypes and patterns | | [agents/tools.md](references/agents/tools.md) | Tool configuration patterns | | [agents/discovery.md](references/agents/discovery.md) | How agents are found and loaded | | [agents/patterns.md](references/agents/patterns.md) | Best practices, multi-agent patterns | | [agents/tasks.md](references/agents/tasks.md) | Task tool patterns | | [agents/task-tool.md](references/agents/task-tool.md) | Task tool integration | | [agents/performance.md](references/agents/performance.md) | Performance optimization | | [agents/advanced-features.md](references/agents/advanced-features.md) | Resumable agents, CLI config | | [agents/agent-vs-skill.md](references/agents/agent-vs-skill.md) | Agents vs Skills distinction | --- ## Commands Custom slash commands — reusable prompts invoked by users. ### Frontmatter ```yaml --- description: Deploy to environment with validation # Required argument-hint: <environment> [--skip-tests] # Shown in autocomplete allowed-tools: Bash(*), Read # Restrict tool access model: claude-3-5-haiku-20241022 # Override model --- ``` See [commands/frontmatter.md](references/commands/frontmatter.md) for complete schema. ### Core Features <!-- <bang> = exclamation mark; used as stand-in to avoid triggering preprocessing --> | Feature | Syntax | Purpose | | ----------------- | ------------------------ | ------------------------------- | | Arguments | `$1`, `$2`, `$ARGUMENTS` | Dynamic input from user | | Bash execution | `` <bang>`command` `` | Include shell output in context | | File references | `@path/to/file` | Include file contents | | Tool restrictions | `allowed-tools:` | Limit Claude's capabilities | ### Quick Example ```markdown --- description: Fix a specific GitHub issue argument-hint: <issue-number> --- Fix issue #$1 following our coding standards. Review the issue, implement a fix, add tests, and create a commit. ``` Usage: `/fix-issue 123` ### Argument Patterns - **Positional**: `$1`, `$2`, `$3` — `/compare old.ts new.ts` - **All arguments**: `$ARGUMENTS` — `/fix memory leak in auth` - **With file refs**: `@$1` — `/analyze src/main.ts` See [commands/arguments.md](references/commands/arguments.md) for advanced patterns. ### Scopes | Scope | Location | Shows in /help | | -------- | --------------------- | -------------- | | Project | `.claude/commands/` | "(project)" | | Personal | `~/.claude/commands/` | "(user)" | | Plugin | `<plugin>/commands/` | Plugin name | Subdirectories create namespaces: `commands/frontend/component.md` → `/component`. ### Validation Checklist - [ ] Filename is kebab-case `.md` (verb-noun pattern preferred) - [ ] `description` is action-oriented, under 80 chars - [ ] `argument-hint` uses `<required>` and `[optional]` syntax - [ ] `allowed-tools` uses correct case-sensitive names - [ ] Bash commands tested in terminal first ### References | Reference | Content | | --------------------------------------------------------------------- | --------------------------- | | [commands/frontmatter.md](references/commands/frontmatter.md) | Complete frontmatter schema | | [commands/arguments.md](references/commands/arguments.md) | Argument handling patterns | | [commands/bash-execution.md](references/commands/bash-execution.md) | Shell command execution |