claudit

# Claudit: Claude Code Configuration Audit

You are the Claudit orchestrator. When the user runs `/claudit` or `/claudit <focus-area>`, execute this 6-phase audit workflow (Phases 0–5). Follow each phase in order. Do not skip phases.

## Phase 0: Environment Detection & Configuration Map

### Step 1: Environment Detection

1. **PROJECT_ROOT**: Run `git rev-parse --show-toplevel 2>/dev/null` via Bash. If this fails (not in a git repo), set PROJECT_ROOT to empty.
2. **HOME_DIR**: Run `echo $HOME` via Bash.

### Step 1.5: Parse Focus Argument

Extract the focus area from `$ARGUMENTS`.

- If `$ARGUMENTS` is empty or missing → set **FOCUS_MODE = false**. Proceed with a full audit (default behavior).
- If `$ARGUMENTS` is provided → set **FOCUS_MODE = true**. Set **FOCUS_AREA** to the user's input.

Map `$ARGUMENTS` to a focus area using the **Focus-Area Mapping** table in `${CLAUDE_PLUGIN_ROOT}/skills/claudit/references/report-templates.md`. Store **FOCUS_AREA** (the interpreted label) and **FOCUS_CATEGORIES** (the relevant scoring categories) for use in later phases. Plugin-name matching is deferred to Step 3.5 (it requires reading `installed_plugins.json`); apply only keyword matching here.

### Step 2: Scope Detection

- If `PROJECT_ROOT` is found → **comprehensive** (global + project)
- If `PROJECT_ROOT` is empty → **global only**

### Step 3: Comprehensive Configuration Scan

Run parallel Glob calls to discover every Claude-related file. Cap at 50 total files — if a project has more, report the cap and proceed with the 50 most recently modified.

**Project-level (if comprehensive):**

| Category | Glob Pattern | Notes |
|----------|-------------|-------|
| Instructions | `{PROJECT_ROOT}/**/CLAUDE.md` | Exclude node_modules, .git, vendor, dist, build via pattern |
| Local instructions | `{PROJECT_ROOT}/CLAUDE.local.md` | Personal/gitignored |
| Rules | `{PROJECT_ROOT}/.claude/rules/**/*.md` | Modular rules with optional path frontmatter |
| Settings (shared) | `{PROJECT_ROOT}/.claude/settings.json` | Team settings |
| Settings (local) | `{PROJECT_ROOT}/.claude/settings.local.json` | Personal project settings |
| Skills | `{PROJECT_ROOT}/.claude/skills/*/SKILL.md` | Project skills |
| Agents | `{PROJECT_ROOT}/.claude/agents/*.md` | Project subagents |
| Memory | `{PROJECT_ROOT}/.claude/MEMORY.md` | Project memory |
| MCP | `{PROJECT_ROOT}/.mcp.json` | Project MCP servers |
| Plugin hooks | `{PROJECT_ROOT}/.claude/plugins/*/hooks/hooks.json` | Plugin-level hooks |

For the Instructions glob, exclude common vendor directories. Use Glob with pattern `**/CLAUDE.md` rooted at PROJECT_ROOT, then filter out paths containing `node_modules`, `.git`, `vendor`, `dist`, or `build`.

**Global-level (always):**

| Category | Path | Notes |
|----------|------|-------|
| Settings | `~/.claude/settings.json` | Global settings |
| Instructions | `~/.claude/CLAUDE.md` | Global instructions (check `~/CLAUDE.md` too as legacy) |
| Rules | `~/.claude/rules/**/*.md` | Personal modular rules |
| Memory | `~/.claude/MEMORY.md` | Global memory |
| MCP | `~/.claude/.mcp.json` | Global MCP servers |
| Plugins | `~/.claude/plugins/installed_plugins.json` | Installed plugins |
| Marketplaces | `~/.claude/plugins/known_marketplaces.json` | Registered marketplaces |
| Managed policy (macOS) | `/Library/Application Support/ClaudeCode/CLAUDE.md` | macOS managed policy |
| Managed policy (Linux) | `/etc/claude-code/CLAUDE.md` | Linux/WSL managed policy |

For each file found, get its line count via `wc -l` (batch multiple files in a single Bash call for efficiency). Quote paths containing spaces (e.g., `/Library/Application Support/...`) in any Bash commands.

### Step 3.5: Resolve Plugin Name Focus

If **FOCUS_MODE is true** and **FOCUS_AREA** is still "Free-form": check whether `$ARGUMENTS` matches an installed plugin name. Read `installed_plugins.json` (if it was found in the config scan) and check whether `$ARGUMENTS` is an exact match or substring of any plugin name. If so, reclassify: set **FOCUS_AREA** to "Specific Plugin: {name}" and **FOCUS_CATEGORIES** to Plugin Health.

### Step 4: Build and Present the Configuration Map

Build a structured manifest grouping files by category with line counts, then present it (and, if **FOCUS_MODE is true**, the focus summary) following the **Configuration Map** template in `${CLAUDE_PLUGIN_ROOT}/skills/claudit/references/report-templates.md`. That template also defines the instruction-file token estimate to display.

### Step 5: Load Decision Memory

Load past audit decisions so they can be injected into audit agents and annotated in the report. Follow the read procedure in `${CLAUDE_PLUGIN_ROOT}/references/decision-memory-protocol.md`.

1. Determine the decisions file path:
   - **Comprehensive scope**: `{PROJECT_ROOT}/.claude/claudit-decisions.json`
   - **Global only**: `~/.cache/claudit/decisions.json`
2. Read the file via Bash: `cat {path} 2>/dev/null`
3. If it exists and parses as valid JSON with `schema_version: 1` → store `decisions` array as **DECISION_HISTORY**
4. If it doesn't exist → set **DECISION_HISTORY** to empty array (first run)
5. Add to the config map display: `Decision Memory: N past decisions` (or `Decision Memory: none (first run)`)

Also run `git config user.name 2>/dev/null` and store as **GIT_USER** for use in Phase 4.

Then tell the user:

```
Phase 1: Building expert context from official Anthropic documentation...
```

---

## Phase 1: Build Expert Context

### Step 1: Load Expert Context

Invoke `/claudit:knowledge all` to retrieve all three knowledge domains (core-config, ecosystem, optimization). The knowledge skill checks cache freshness and auto-refreshes stale domains.

Use the output as **Expert Context** for Phase 2.

Tell the user:
```
Phase 2: Analyzing your configuration against expert knowledge...
```

---

## Phase 2: Expert-Informed Audit

Dispatch audit subagents using the Task tool. Each agent receives the **Expert Context** from Phas