agentleFS
Sign inStart free
MegapromptingMegaprompting/torque-loop/skills/attack/SKILL.md

attack

Skill10 starsChanged 4 months agoMITView on GitHub

Run adversarial review on a finished-looking artifact. Use to replace self-praise with hostile pressure before shipping. Convenes a five-voice review board (impatient user, competitor, maintainer, auditor, saboteur), returns concrete failure modes with severity and evidence demanded, and the smallest patch each requires.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. jantichy/claude★ 45same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository24 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    If nothing critical or high survived a genuine attack, say so plainly — that is a real result.

    line 71

≈ “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

/ratchet:attack — hostile validation

Self-review is theater. The model that built the artifact is the worst judge of it. This command assumes the artifact is wrong and spends its effort proving where.

Step 0 — Load state and target

ratchet status

Identify the artifact under attack (usually the last one in state). For anything that looks finished, delegate the assault to the ratchet-auditor subagent.

The five-voice board

Speak in each voice distinctly. Do not blur them into generic "concerns".

  1. Impatient User — slower, harder, or more confusing than the alternative.
  2. Competitor — where a rival beats this and takes the user.
  3. Maintainer — how this rots or becomes unchangeable in six months.
  4. Auditor — which claims are asserted without evidence.
  5. Saboteur — the specific input or state that makes it fall over.

Rules

  • Every objection is a concrete failure scenario: specific input/state → wrong output or crash. "Might be fragile" is not a finding; "empty input throws at step 3" is.
  • Rank by severity: critical / high / medium / low. Critical = wrong result, data loss, or unusable. Order most-severe first.
  • Demand evidence for every self-serving claim ("robust", "fast", "complete"). Find the contradicting case or mark the claim unproven.
  • Name the smallest patch each finding needs — REMOVE / ADD / CHANGE. You specify the delta; you do not apply it here.
  • Flag fog, not just defects. A finding that says the premise was wrong — a hidden convention, wrong-by-default data, a claim that contradicts the locked target or the unknowns-map — is fog the aperture missed, not merely a bug. Serialize it (ratchet state append assumptions ... with a kill test, or an openLoops entry prefixed fog:), and at two or more such findings recommend /ratchet:map before any further patching: you are patching inside unmapped terrain.

Output contract

VERDICT: ship | patch-then-ship | do-not-ship
FINDINGS (most severe first):
- [CRITICAL] <voice> — <failure scenario> → <REMOVE/ADD/CHANGE ...>
- [HIGH] ...
UNPROVEN CLAIMS:
- "<quote>" — evidence demanded: ...

Serialize

Record every real defect (one call each) so confidence reflects reality:

ratchet defect add '{"severity":"critical","summary":"...","artifact":"<id>"}'
ratchet state set phase attack

Name the artifact the defect attacks. With exactly one live artifact the CLI attaches it for you; with two or more it refuses rather than guess, because attaching to the wrong one blocks the innocent artifact's closure and lets the guilty one close clean.

If nothing critical or high survived a genuine attack, say so plainly — that is a real result. Next: /ratchet:patch to fix exactly what failed.

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.

Check your setup, free