Brewdoc:My Claude

My Claude

Generates documentation about your Claude Code installation and environment.

Mode Detection

Detect mode from $ARGUMENTS:

After detection, load the appropriate reference file:

• INTERNAL: references/internal-mode.md
• EXTERNAL: references/external-mode.md
• RESEARCH: references/research-mode.md

vs built-in /team-onboarding

Claude Code 2.1.101 shipped /team-onboarding — a built-in slash that generates a teammate ramp-up guide from local usage. It's simpler and sufficient for basic onboarding.

Use /team-onboarding when you need: quick teammate handoff doc from your local config.

Use /brewdoc:my-claude when you need:

• Web research — current releases, forum discussions, GitHub issues (beyond static docs)
• EXTERNAL mode — architecture synthesis from official Anthropic docs
• RESEARCH mode — custom query-driven multi-source investigation
• Persistent INDEX — tracked in .claude/brewdoc/INDEX.jsonl with citation links

Output Directory

All generated docs go to .claude/brewdoc/my-claude/ (project-relative — required because ~/.claude/* is blocked by Claude Code's protected-path policy in headless sessions, even under bypassPermissions). Create if not exists: mkdir -p .claude/brewdoc/my-claude

Optional interactive fallback: ${BD_PLUGIN_DATA}/my-claude/ may be used when running interactively — it is NOT usable in headless claude -p sessions due to the protected-path restriction. Prefer the project-relative path everywhere.

INDEX Tracking

Append entry to .claude/brewdoc/INDEX.jsonl:

{"ts":"2026-02-28T10:00:00","mode":"internal","path":".claude/brewdoc/my-claude/20260228_my-claude-internal.md","title":"Internal Claude Setup Overview","version":"1.0"}

Legacy read-only merge — if ~/.claude/brewdoc/INDEX.jsonl exists AND the new project INDEX is empty, read the legacy file once, merge its entries into .claude/brewdoc/INDEX.jsonl, and print: ℹ️ Migrated {N} entries from legacy ~/.claude/brewdoc/INDEX.jsonl (read-only; legacy file untouched). NEVER write back to the legacy path.

If an existing entry for the same mode exists: use AskUserQuestion — header: "INDEX", question: "Entry for this mode already exists (v{VERSION}). Update it?", options: "Yes, update (bump version)" / "No, create new entry".

INTERNAL Mode

Goal: Document your local Claude Code setup — CLAUDE.md files, rules, agents, skills, memories.

Sources to analyze:

• ~/.claude/CLAUDE.md — global instructions
• ~/.claude/rules/*.md — global rules
• ~/.claude/agents/*.md — global agents
• ~/.claude/skills/ — global skills
• Project CLAUDE.md (current working directory)
• .claude/rules/*.md — project rules
• ~/.claude/projects/**/memory/MEMORY.md — memory files

Process: 1. Spawn 3 parallel Explore agents, one per source group: (1) global ~/.claude config, (2) project .claude config, (3) memory files 2. Aggregate findings into structured document 3. Write to .claude/brewdoc/my-claude/YYYYMMDD_my-claude-internal.md 4. Spawn independent reviewer agent to validate facts (file paths exist, content accurate) 5. Apply reviewer fixes if any 6. Add INDEX entry

Output document structure:

# Claude Code Internal Setup — {date}

## Global Configuration
### Instructions (CLAUDE.md)
### Rules ({N} rules)
### Agents ({N} agents)
### Skills ({N} skills)

## Project Configuration
### Project Instructions
### Project Rules

## Memory
### Active Memories ({N} entries)

## Summary
| Component | Count | Location |
|-----------|-------|----------|

EXTERNAL Mode

Goal: Document Claude Code's hook/context/agent architecture from official sources + local analysis.

Sub-mode default: 1. Analyze local hook files for event model patterns 2. WebSearch for recent Claude Code releases and CHANGELOG 3. Spawn general-purpose agents for: official docs (code.claude.com), GitHub releases, community forums 4. Generate .claude/brewdoc/my-claude/YYYYMMDD_my-claude-external.md

Sub-mode context-schema: 1. Focus specifically on context injection schema (additionalContext, updatedInput, etc.) 2. Output: .claude/brewdoc/my-claude/external/YYYYMMDD_context-schema.md

RESEARCH Mode

Goal: Research a specific query about Claude Code using multiple sources.

Query: everything after r or research in $ARGUMENTS

Process: 1. Analyze query — divide into 2-5 source groups (official docs, GitHub, Reddit, forums, marketplaces) 2. Spawn general-purpose agents per source group in parallel 3. Aggregate with citation tracking (source URL per fact) 4. Spawn independent reviewer agent to validate facts and source reliability 5. Output: .claude/brewdoc/my-claude/YYYYMMDD_research-{slug}.md

Output structure:

# Research: {query} — {date}

## Findings

### {Source Group 1}
...

## Sources
| Fact | Source | Reliability |
|------|--------|-------------|

## Review Verdict

My Claude

Generates documentation about your Claude Code installation and environment. Supports four modes: document your local setup, analyze the external hook/context architecture, deep-dive the context injection schema, or research any Claude Code topic across the web.

Quick Start

/brewdoc:my-claude              # Document your local Claude Code setup
/brewdoc:my-claude ext          # Document hook/context/agent architecture
/brewdoc:my-claude ext context  # Deep-dive into context injection schema
/brewdoc:my-claude r <query>    # Research a specific Claude Code topic

Modes

Examples

Good Usage

# Document everything in your local Claude Code installation
/brewdoc:my-claude

# Generate docs on hook events, agent architecture, and recent releases
/brewdoc:my-claude ext

# Understand how additionalContext and updatedInput work in hooks
/brewdoc:my-claude ext context

# Research how other projects structure their plugin hooks
/brewdoc:my-claude r plugin hook patterns and best practices

# Find out what changed in recent Claude Code versions
/brewdoc:my-claude r Claude Code changelog 2026

Common Mistakes

# Wrong: using research mode for a code implementation task
/brewdoc:my-claude r implement a new hook for my project
# Research mode generates documentation, not code. Use /brewcode:start for implementation.

# Wrong: using ext when you want to document YOUR setup
/brewdoc:my-claude ext
# ext documents the external Claude Code architecture. Use no arguments for your local setup.

# Wrong: forgetting the space after r
/brewdoc:my-claude rsome query
# The prefix must be "r " or "research " (with a trailing space).

Output Location

All generated files are saved to .claude/brewdoc/my-claude/ inside your current project. This is required because Claude Code's protected-path policy blocks writes to ~/.claude/* in headless sessions (including bypassPermissions mode). ${BD_PLUGIN_DATA}/my-claude/ is an optional fallback for interactive sessions only.

Every run appends an entry to .claude/brewdoc/INDEX.jsonl for tracking. If an entry for the same mode already exists, you will be asked whether to update it or create a new one.

If a legacy ~/.claude/brewdoc/INDEX.jsonl exists from an older install, its entries are merged into the project INDEX once (read-only) and the skill prints a one-line migration notice. The legacy file is never written to.

Tips

• Re-run internal mode after changing your setup. Adding new rules, agents, or skills makes the previous internal doc stale. Re-running produces an up-to-date snapshot.
• Research mode is for documentation, not implementation. It generates a cited research report. For code tasks, use brewcode skills instead.
• External context-schema mode is useful for hook development. Before writing a new hook, run ext context to get a fresh reference on which channels (additionalContext, updatedInput, decision) work for which events.
• Check INDEX.jsonl to find previous runs. Each entry includes timestamp, mode, path, and version, so you can track what was documented and when.

Documentation

Full docs: my-claude

External Mode Reference

Hook Event Model

When analyzing local hook files, document:

• Event types (SessionStart, PreToolUse, PostToolUse, Stop, PreCompact)
• Input schema per event
• Output schema (additionalContext, updatedInput, decision)
• permissionDecision values

Sources for Web Research

Output Structure

# Claude Code External Architecture — {date}

## Hook Event Model
### Events
### Input Schema
### Output Schema

## Context Injection Patterns
## Agent Spawning API
## Recent Changes (CHANGELOG)

## Sources

Context-Schema Sub-mode

Focus on:

• additionalContext — what it does, when delivered, format
• updatedInput — mutation of tool input
• systemMessage — display in UI
• permissionDecision — allow/block/ask

Output to: .claude/brewdoc/my-claude/external/YYYYMMDD_context-schema.md

Internal Mode Reference

Sources to Analyze

Parallel Agent Split

Spawn 3 Explore agents in ONE message: 1. Agent 1: global ~/.claude/ directory (rules, agents, skills counts + key content) 2. Agent 2: project .claude/ directory (project-specific config) 3. Agent 3: memory files (active entries, topic files)

Document Template

# Claude Code Internal Setup — {YYYY-MM-DD}

## Global Configuration

### Instructions
**File:** `~/.claude/CLAUDE.md`
{summary of key rules/frameworks}

### Rules ({N} files)
| Rule File | Purpose |
|-----------|---------|
{table of rule files}

### Agents ({N} files)
| Agent | Model | Purpose |
|-------|-------|---------|

### Skills ({N} files)
| Skill | Trigger | Purpose |
|-------|---------|---------|

## Project Configuration
**File:** `{cwd}/CLAUDE.md`
{project-specific instructions summary}

### Project Rules ({N} files)
{table}

## Memory
### Active Memories
{summary of MEMORY.md content}

### Topic Files
{list of additional memory files}

## Summary
| Component | Count | Location |
|-----------|-------|----------|
| Global rules | {N} | ~/.claude/rules/ |
| Global agents | {N} | ~/.claude/agents/ |
| Global skills | {N} | ~/.claude/skills/ |
| Project rules | {N} | .claude/rules/ |
| Memory files | {N} | ~/.claude/projects/.../ |

Reviewer Checklist

• File paths referenced in document actually exist
• Agent/skill counts match filesystem counts
• No invented file names

Research Mode Reference

Query Partitioning Strategy

Divide query into 2-5 orthogonal source groups:

Parallel Agent Spawn

Send all agents in ONE message. Each agent receives:

• Source group name and URLs to search
• Specific aspects to focus on
• Citation format requirement

Citation Format

Each fact must have: [Source: URL] (reliability: high/medium/low)

Reliability:

• high: Official docs, official changelog
• medium: GitHub issues, verified community posts
• low: Reddit opinions, unverified claims

Output Structure

# Research: {query} — {date}

## Executive Summary

## Findings by Source

### Official Documentation
{findings with citations}

### GitHub
{findings with citations}

### Community
{findings with citations}

## Consolidated Facts
| Fact | Source | Reliability |
|------|--------|-------------|

## Conflicting Information
{if any}

## Review Verdict
{reviewer agent assessment}

## Sources
{full URL list}

Self-Review Checklist (for reviewer agent)

• All facts have citations
• No contradictions between high-reliability sources
• Conflicting low-reliability claims flagged
• Query fully addressed