agentleFS
r-irber-irbe/proof-skills/skills/lean-tautology-triage/SKILL.md

lean-tautology-triage

Skill3 starsChanged 27 days agoApache-2.0View on GitHub

USE FOR: triaging Lean 4 theorem statements or proof sites that may be vacuous, tautological, smoke-test-only. DO NOT USE FOR: ordinary proof writing (@lean-proof). TRIGGERS: tautology, vacuous, trivial proof, smoke theorem.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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. Distinctive:

    STOP if the statement could be SUSPECT-FALSE or UNVERIFIED and confidence is below the repo belief floor.

    line 47
  2. Distinctive:

    STOP before deleting or weakening a public theorem without an explicit edit claim and rollback path.

    line 49
  3. Very specific:

    STOP if static proof_quality.py output conflicts with manual inspection.

    line 51
  4. Distinctive:

    STOP if a proof is short because of a verified definitional theorem in Mathlib.

    line 53

≈ “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

lean-tautology-triage

This skill is a focused review mode for statements that elaborate but may not

Routing

  • USE FOR: triaging Lean 4 theorem statements or proof sites that may be vacuous, tautological, smoke-test-only, reflexive-by-rfl, : True placeholders, bare decide closures, or automation-only proofs of supposedly substantive claims.
  • DO NOT USE FOR: ordinary proof writing (use @lean-proof); whole-project QA lifecycle (use @lean-quality-engine); running one enforcement script without interpretation (use @lean-enforcement); theorem search (use @lean-research).
  • TRIGGERS: tautology, vacuous, trivial proof, smoke theorem, placeholder

Workflow

  1. Locate the claim [discover] — capture theorem name, statement, proof body, surrounding docstring, imports, and downstream uses (via lean_navigate {"operation": "references", ...} or lean_decl_references with {"symbol": "...", "direction": "usedBy"}). A claim with downstream uses is not automatically substantive, but the use sites explain risk.
  2. Classify statement content [validate] — use:
    • VALID-CONTENT: theorem states the intended mathematical/program property.
    • VALID-SMOKE: theorem intentionally checks elaboration or examples only.
    • VACUOUS: theorem is true because hypotheses are impossible or conclusion is True.
    • MISALIGNED: theorem is true but the name/docstring overclaims.
    • UNVERIFIED: statement may be content-bearing but lacks evidence.

Recovery & STOP

  • STOP if the statement could be SUSPECT-FALSE or UNVERIFIED and confidence is below the repo belief floor; ask before recommending keep/delete.
  • STOP before deleting or weakening a public theorem without an explicit edit claim and rollback path.
  • STOP if static proof_quality.py output conflicts with manual inspection; cite both and ask which risk posture to use.
  • STOP if a proof is short because of a verified definitional theorem in Mathlib;

Handoffs

  • Predecessors: @lean-proof-review for proof-site review, @lean-quality-engine for project-level quality gates, @lean-enforcement for script execution.
  • Successors: @lean-proof-review for final review, @lean-research for theorem/source lookup, @lean-quality-engine for gate rollups.
  • Script: scripts/lean/proof_quality.py surfaces candidates; reviewers own final classification.

Detailed reference

Full encyclopedic reference, decision trees, and lookup tables live in references/lean-tautology-triage-handbook.md .

See also

  • ../../references/lean-tautology-triage-handbook.md -- Complete lean-tautology-triage Handbook
  • ../lean-gateway/SKILL.md -- Top-level skill router

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.