agentleFS
kucherenkokucherenko/gangsta/skills/audit-review/SKILL.md

audit-review

Skill85 starsChanged 20 days agoMITView on GitHub

Use when completing tasks, implementing major features, or before merging — dispatches the-inspector to audit the books before the job is closed

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After each task in parallel execution (The Hit)

    line 17
  2. Distinctive:

    After completing a major feature

    line 18
  3. Distinctive:

    Before merge to main

    line 19
  4. Distinctive:

    When stuck (fresh perspective)

    line 22
  5. Distinctive:

    Before refactoring (baseline check)

    line 23
  6. Distinctive:

    After fixing complex bug

    line 24
  7. Very specific:

    Use the Task tool to dispatch the the-inspector agent with subagent_type: "the-inspector".

    line 37
  8. Very specific:

    Use gangsta:receiving-orders for processing feedback rigorously

    line 99

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

The Audit: Requesting Code Review

Overview

Dispatch the-inspector to catch issues before they compound. The inspector gets precisely crafted context for evaluation — never the session's history. This keeps the inspector focused on the work product, not the thought process, and preserves your context for continued work.

Core principle: Audit early, audit often.

When to Request

Mandatory:

  • After each task in parallel execution (The Hit)
  • After completing a major feature
  • Before merge to main

Optional but valuable:

  • When stuck (fresh perspective)
  • Before refactoring (baseline check)
  • After fixing complex bug

How to Request

1. Get Git SHAs

BASE_SHA=$(git rev-parse HEAD~1)  # or origin/main
HEAD_SHA=$(git rev-parse HEAD)

2. Dispatch the-inspector

Use the Task tool to dispatch the the-inspector agent with subagent_type: "the-inspector". Do NOT use "general" or "general-purpose" — these are not valid in a Gangsta Agents installation.

Fill the template at the-inspector-prompt.md in this skill directory with the following placeholders:

  • {WHAT_WAS_IMPLEMENTED} — What you just built
  • {PLAN_OR_REQUIREMENTS} — What it should do (Contract clause, spec section)
  • {BASE_SHA} — Starting commit
  • {HEAD_SHA} — Ending commit
  • {DESCRIPTION} — Brief summary of the changes

3. Act on Findings

SeverityAction
CriticalFix immediately. Do not proceed until resolved.
ImportantFix before proceeding to the next task.
MinorNote for later. Don't block progress.
WrongPush back with technical reasoning.

Example

[Just completed implementing auth middleware]

1. Get SHAs:
   BASE_SHA=$(git log --oneline | grep "previous task" | head -1 | awk '{print $1}')
   HEAD_SHA=$(git rev-parse HEAD)

2. Dispatch the-inspector:
   WHAT_WAS_IMPLEMENTED: JWT auth middleware with role-based access
   PLAN_OR_REQUIREMENTS: Contract section 3.2 — Authentication
   BASE_SHA: a7981ec
   HEAD_SHA: 3df7661
   DESCRIPTION: Added auth middleware, role guards, token validation

3. Inspector returns:
   Strengths: Clean separation, real tests
   Issues:
     Important: Missing token expiry check
     Minor: Magic number for token TTL
   Assessment: Ready with fixes

4. Fix important issue, proceed to next task.

Integration with Heist Pipeline

  • The Hit: Audit after each Crew Lead's territory completion
  • Laundering: Final audit before the Don's approval
  • Ad-hoc work: Audit before merge

Red Flags

Never:

  • Skip audit because "it's simple"
  • Ignore Critical issues
  • Proceed with unfixed Important issues
  • Argue with valid technical feedback without evidence

If inspector is wrong:

  • Push back with technical reasoning
  • Show code/tests that prove it works
  • Use gangsta:receiving-orders for processing feedback rigorously

Omerta Compliance

  • Rule of Truth: Inspector reviews actual diffs, not claims
  • Introduction Rule: Inspector communicates through the audit skill, not directly with Workers

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.