agentleFS
aradmanamnaoonaradmanamnaoon/vibe-coding-skills/skills/agent-reliability/completion-claims-auditor/SKILL.md

completion-claims-auditor

Skill0 starsChanged 19 days agoMITView on GitHub

Audit an agent's claim that work is complete by mapping each claim to observed repository, test, UI, API, or remote evidence. Use before saying done, shipped, fixed, pushed, or verified.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read applicable repository instructions before edits.

    line 20
  2. Distinctive:

    Check that no failing/skipped required gate was hidden or weakened.

    line 38
  3. Distinctive:

    Write the final status using only evidence actually observed in this session/repository.

    line 39
  4. Distinctive:

    Never manufacture screenshots, command results, CI status, remote publication, or test outcomes.

    line 47
  5. Distinctive:

    Every completion statement points to evidence or is explicitly qualified.

    line 51
  6. Distinctive:

    If a required evidence source is inaccessible, report the implementation plus the missing verification and the concrete next check.

    line 64

≈ “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

Purpose

Make completion an evidence state rather than a language-model judgment.

When to activate

An agent is about to report a multi-step coding task, feature, fix, release, push, or deployment as complete.

When not to activate

A purely advisory response has no implementation or execution claim.

Preconditions

  • Read applicable repository instructions before edits.
  • Preserve unrelated working-tree changes and establish a baseline when the workflow can modify code.
  • Identify which tools and environments are actually available; do not claim unavailable checks ran.

Required inputs

  • Original user requirements
  • Current diff/repository state
  • Test/tool outputs
  • Remote state when relevant

Workflow

  1. Extract every externally meaningful completion claim from the requirements and current work.
  2. For each claim, name the strongest evidence type that would establish it: file state, compilation, test, browser flow, API response, artifact smoke, Git remote hash, CI result, or external state.
  3. Collect/execute that evidence and record the observed result; do not substitute intent or code inspection when runtime evidence is feasible.
  4. Distinguish implemented, structurally validated, functionally verified, externally verified, and not executed.
  5. Reopen any claim contradicted by evidence and fix or downgrade it before reporting.
  6. Check that no failing/skipped required gate was hidden or weakened.
  7. Write the final status using only evidence actually observed in this session/repository.

Relevant tools

Repository state, test/build tools, browser/API checks, Git/GitHub/CI state.

Safety restrictions

Never manufacture screenshots, command results, CI status, remote publication, or test outcomes.

Validation

  • Every completion statement points to evidence or is explicitly qualified.
  • Local commit is not called pushed without remote verification.
  • Code existence is not called working when required runtime checks were not executed.

Expected output

  • Claim-to-evidence ledger
  • Failed/missing evidence
  • Final allowed wording
  • Exact blockers

Failure handling

If a required evidence source is inaccessible, report the implementation plus the missing verification and the concrete next check; do not invent a percentage.

Completion criteria

The skill is complete only when every applicable validation item above has observed evidence or is explicitly marked not executed with the concrete blocker. Do not replace missing evidence with confidence language.

Research mapping

This skill was selected from research problems: P044, P039. See research/vibe-coding-problems.md and research/problem-evidence.csv in the Vibe Coding Skills repository.

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.