agentleFS
excattexcatt/superclaude-plusplus/skills/audit/SKILL.md

audit

Skill14 starsChanged 37 days agoMITView on GitHub

Run project-specific verification rules (business logic, architecture patterns, naming) that linters cannot catch; bootstraps rules if none exist. Not for build/lint/test (/verify).

194 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use teach-impeccable, arrange and 3 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • HK-hubSkill

    teach-impeccable

    One-time setup that gathers design context for your project and saves it to your AI config file. Run once to establish persistent design guidelines.

    Also in 19 projects that use this

  • HK-hubSkill

    arrange

    Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy. Use when the user mentions layout feeling off, spacing issues, visual hierarchy, crowded UI, alignment problems, or wanting better composition.

    Also in 14 projects that use this

  • zwy123zwySkill

    normalize

    将用户随意输入的提示词规范化为结构清晰、便于大模型执行的标准化提示词。适用于需要整理、统一提示词格式时调用。

    Also in 19 projects that use this

  • SkillPilotCHSkill

    level-up

    Eine Automatisierung pro Woche finden, scopen, shippen. Output landet als proposed-Eintrag in cadence.md mit konkreten Implementations-Schritten.

    Also in 12 projects that use this

  • ww-w-aiSkill

    pm-discovery

    PM Agent Team - Automated product discovery, strategy, and PRD generation. Runs PM analysis workflow to produce comprehensive PRD before PDCA Plan phase.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 2 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 175 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Do NOT use for: build/lint/test (/verify), design vs implementation (/gap-analysis)

    line 21
  2. Very specific:

    Read all .claude/audit-rules/*.md.

    line 39
  3. Very specific:

    Test fixtures: fixtures/, __fixtures__/, test-data/

    line 53
  4. Very specific:

    Run bash commands from ## Check section (if any)

    line 61
  5. Very specific:

    Ask user via AskUserQuestion:

    line 84
  6. Distinctive:

    After fixes, re-run only failed rules and show before/after comparison.

    line 89
  7. Distinctive:

    Always confirm before creating or deleting rules:

    line 116

“Common” is measured: kept by at least 6 in 10 of the 175 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Audit Skill

Project-specific verification rules that generic linters cannot enforce.

SkillChecks
/verifyBuild, types, lint, test, security
/auditBusiness logic, architecture patterns, naming conventions
/gap-analysisDesign doc vs implementation matchRate

When to Use

Triggers: audit, check rules, run audit, audit manage Do NOT use for: build/lint/test (/verify), design vs implementation (/gap-analysis)

Mode Selection

/audit          → rules exist? → RUN mode (execute rules)
                → no rules?   → MANAGE mode (bootstrap)
/audit manage   → MANAGE mode (create/delete/sync rules)

Rules location: .claude/audit-rules/*.md


RUN Mode

Step 1: Collect & Match

Read all .claude/audit-rules/*.md. Match each rule's scope against changed files:

BASE=$(git remote show origin 2>/dev/null | grep 'HEAD branch' | awk '{print $NF}')
BASE=${BASE:-main}
git diff "$BASE"...HEAD --name-only

Skip rules whose scope has no matching changes.

Auto-exempt files (always skip, no rule needed):

  • Lock files: pnpm-lock.yaml, uv.lock, package-lock.json, Cargo.lock
  • Generated/build output: dist/, build/, .next/, __pycache__/
  • Docs: README.md, CHANGELOG.md, LICENSE
  • Test fixtures: fixtures/, __fixtures__/, test-data/
  • Vendor/third-party: vendor/, node_modules/
  • CI/CD config: .github/, .gitlab-ci.yml, Dockerfile
  • CLAUDE.md and .claude/ internals

Step 2: Execute

For each matched rule:

  1. Run bash commands from ## Check section (if any)
  2. AI-analyze results against ## Pass Criteria
  3. Classify: PASS / FAIL / WARN / SKIP

Step 3: Report

AUDIT REPORT
=============
  api-response-format      PASS
  error-handling           FAIL  2 violations
    src/api/users.ts:45    bare catch without error type
    src/api/posts.ts:78    empty catch block
  naming-convention        PASS
---------------------------------------------
  Rules: 3 | Passed: 2 | Failed: 1
  Status: NEEDS FIX

Status: all pass → CLEAR / any error-severity fail → NEEDS FIX / warning-only → REVIEW

Step 4: Fix (if issues found)

Ask user via AskUserQuestion:

  1. Apply All Fixes — Automatically apply all recommended fixes
  2. Review Each Fix — Approve each fix individually
  3. Skip Fixes — Report only, make no changes

After fixes, re-run only failed rules and show before/after comparison.


MANAGE Mode

Bootstrap (No Rules)

Detect project type from config files and propose seed rules:

SignalTypeSuggested Rules
next in package.jsonNext.jsServer/Client separation, API response format
fastapi in pyproject.tomlFastAPIHTTPException usage, router naming
express/nest in package.jsonNode APIError middleware, response wrapper
Any projectUniversalNaming convention

Change Analysis

git diff "$BASE"...HEAD --name-only --stat

Compare changed domains against existing rule scopes → propose CREATE/UPDATE/DELETE.

User Confirmation

Always confirm before creating or deleting rules:

Proposed:
  + api-response-format   (new, covers src/api/**)
  - legacy-auth-check     (stale, files removed)

Apply? [Y/n]

Rule Format

Each rule is a markdown file with YAML frontmatter (name, scope, severity) and structured sections (What, Check, Pass Criteria, Examples).

Full guide: references/rule-format.md Example rules: references/examples/


Integration

Pre-commit flow: /verify → /audit → commit

Review loop: Implement → Spec Review → Quality Review → /verify → /audit → Complete

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.