feat(plugin): command metadata, scoped skills, local-tier support, layered hooks, Stop audit

Wave 3 - adoption hardening. Patterns adapted (in original prose, with
attribution) from MIT-licensed shanraisshan/claude-code-best-practice.

Commands (command/enhance-claude-md.md, command/sync-claude-md.md):
- Add allowed-tools / disallowedTools / argument-hint / when_to_use so the
  commands auto-suggest in the slash menu and avoid permission prompts.
- disallowedTools blocks WebFetch + WebSearch on both commands.
- Drop the previous broken hooks block (array-of-{matcher, commands} shape
  did not match canonical schema; was never firing).

Skills:
- skill/karpathy-guidelines/SKILL.md: paths: glob over 23 code-file
  extensions, so the guardrails auto-load only when editing source, not
  markdown or data.
- skill/SKILL.md: model: haiku, effort: medium, paths: scoped to CLAUDE.md
  + AGENTS.md + .claude/rules/*.md so validator/generator passes run
  cheaply without changing the user-facing model.

CLAUDE.local.md personal tier:
- skill/validator.py BestPracticesValidator now accepts filename=; any
  *.local.md basename waives the 150-line cap.
- hooks/validate-claude-md.py reads the exempt suffix from hooks-config.
- .gitignore covers CLAUDE.local.md, **/CLAUDE.local.md,
  .claude/settings.local.json, hooks/hooks-config.local.json.

Layered hook config:
- hooks/hooks-config.json: committed defaults
  (validateClaudeMd.enabled/maxLines/exemptFilenameSuffix/exitCodeOnViolation,
  stopAuditLine.enabled).
- hooks/validate-claude-md.py merges hooks-config.json +
  hooks-config.local.json key-by-key; honours enabled=false (silent
  exit 0), configurable cap, configurable exit code.

Stop audit hook:
- hooks/audit-claude-md.py walks the project tree, prints one stderr
  line: total tracked / OVER cap / near cap (>=80%). Respects
  stopAuditLine.enabled from config.
- hooks/hooks.json registers Stop event with matcher "".

Guardian fail-closed contract:
- agent/claude-md-guardian.md Safety & Validation section now explicitly
  requires Skill-tool invocation (no inline paraphrase of SKILL.md),
  abort on missing validated output, never auto-commit, and respect
  local hook config.

Verified (8/8 smoke tests):
- Both commands parse with new fields and no broken hooks block.
- karpathy paths: 23 globs, includes .py/.ts/.go/.rs.
- skill model=haiku effort=medium with CLAUDE.md path scope.
- Validator: *.local.md (300 lines) -> pass; CLAUDE.md (300) -> fail;
  legacy ctor without filename -> default behavior preserved.
- hooks-config.json valid; validateClaudeMd.enabled=true, maxLines=150.
- Hook validator: default rc=2 on bloated, rc=0 when local override
  disables it, rc=0 on *.local.md (exempt).
- Stop hook entry present; audit script: rc=0 with "5 CLAUDE.md tracked".
- Regression: large-fullstack root still 52 lines with chain imports.

agent/claude-md-guardian.md

@@ -241,8 +241,14 @@ The slash command can invoke me:
 
 ## Safety & Validation
 
-**Critical Validation Rule**:
-"Always validate your output against official native examples before declaring complete."
+**Fail-closed contract** (non-negotiable):
+
+- I invoke `claude-md-enhancer` exclusively through the **Skill tool**. I never read its `SKILL.md` body and act on a paraphrase of it — paraphrase drift is the most common silent-degradation mode for auto-CLAUDE.md tooling.
+- If the skill returns no validated output (missing required sections, validator status ≠ `pass`, or any thrown exception), I **abort the run** and leave the existing CLAUDE.md tree untouched. Partial writes are worse than stale documentation.
+- I never commit on my own. Every change lands in the working tree only; the user reviews `git diff` and chooses when to commit.
+- I respect `hooks/hooks-config.local.json`. If a developer has disabled the validator locally, I treat the cap as advisory for that machine but still warn on the Stop hook.
+
+**Critical Validation Rule**: validate every emitted file against the reference templates in `skill/examples/` and the canonical schema before declaring success.
 
 **My validation checklist**:
 - ✅ Project Structure diagram present
@@ -250,6 +256,8 @@ The slash command can invoke me:
 - ✅ Architecture section reflects actual patterns
 - ✅ Tech Stack lists all major dependencies
 - ✅ Common Commands match package.json scripts
+- ✅ Every emitted CLAUDE.md ≤ 150 lines (cap waived only for `*.local.md`)
+- ✅ Every sub-CLAUDE.md back-links to root; root has matching `@`-imports
 
 ## Installation