agentleFS
Badhon-00Badhon-00/MDCP/skills/mdcp-self-check/SKILL.md

mdcp-self-check

Skill2 starsChanged 5 days agoMITView on GitHub

Run before saying a coding task is done, and when something fails. Self-check (build, types, tests, scanners), defense review (OWASP 2025 abuse cases), and a self-debug loop that stops after three failed attempts and reports. Use for "check your work", "is it safe", "debug this", "why does it fail".

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run what the project has and record each result as pass, fail or not run:

    line 9
  2. Distinctive:

    Add a regression test.

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Self-check, defense and self-debug

1. Self-check (before "done")

Run what the project has and record each result as pass, fail or not run:

  1. npx tsc --noEmit, lint, tests, npm run build
  2. npm audit --omit=dev; gitleaks detect and semgrep --config auto if installed
  3. git diff: no secrets, no unrelated edits, no leftover debug output
  4. The requested behavior, exercised once (run it, call the route, open the page)

2. Defense review

For every new route, action or form, answer each with evidence from the code or a test:

  • Who may call it, and is ownership of the object checked? (A01)
  • Is input validated and output minimal? (A05, data leakage)
  • What happens on timeout, bad input, missing record, two requests at once? (A10)
  • Is it rate-limited, and is it logged without secrets? (A07, A09)
  • Does it add a dependency, and has that dependency been reviewed? (A03) Full list: master-developer/references/security-check.md, web-analysis-shield.md, exploit-sandbox.md. Test only your own code.

3. Self-debug

  1. Reproduce with a failing test or exact steps. Read the whole error.
  2. Form one hypothesis, find the root cause, grep every caller before touching shared code.
  3. Fix once, in the shared place. Re-run the failing test, then the suite and build.
  4. A fix that does not work gets reverted, then try a different approach. After three failed attempts, stop and report: error, what was tried, what was learned, suggested next step.
  5. Add a regression test.

4. Report

End with: what changed, results of section 1 (including "not run"), open risks, shortcuts taken. Never write "100% secure" or "no bugs".

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.