agentleFS
Sign inStart free
r-irber-irbe/proof-skills/skills/lean-replay-verification/SKILL.md

lean-replay-verification

Skill3 starsChanged 25 days agoApache-2.0View on GitHub

USE FOR: verifying that completed proofs are genuinely kernel-checked after `sorry` removal. DO NOT USE FOR: writing or fixing the proof itself (@lean-proof). TRIGGERS: sorry removed, proof completed, replay, fresh rebuild.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run R1 (cold full build); capture pin, job count, rc into the commit message or ledger.

    line 23
  2. Distinctive:

    Run R2 against the expected-axioms allowlist.

    line 25
  3. Distinctive:

    Check R3 status: is a newer pin available?

    line 26
  4. Very specific:

    If R1 (lake clean + lake build) fails on a tree that passed incrementally, treat the proof as unverified and hand back to @lean-proof or …

    line 33
  5. Very specific:

    If R2 surfaces sorryAx or an unexpected axiom in the closure, block acceptance and route to @lean-axiom-closure / @lean-proof.

    line 37

≈ “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

Replay Verification After Sorry Removal

Operational dispatch contract for @lean-replay-verification.

Routing

  • USE FOR: tasks routed to @lean-replay-verification.
  • DO NOT USE FOR: out-of-scope tasks (delegate via @lean-gateway).

Workflow

  1. Identify the proof-completion event (sorry removed / new theorem accepted as load-bearing).
  2. Run R1 (cold full build); capture pin, job count, rc into the commit message or ledger.
  3. Run R2 against the expected-axioms allowlist; quote the verdict.
  4. Check R3 status: is a newer pin available? Run it via the project's migration flow or record the deferral.
  5. For load-bearing batches, run R4 with trepplein or lean4lean and record checker + version + lineage caveat.
  6. Link the replay evidence from the review record (commit, ledger, or review file) so a later reader can re-derive it.

Recovery & STOP

  • If R1 (lake clean + lake build) fails on a tree that passed incrementally, treat the proof as unverified and hand back to @lean-proof or @lean-build to fix the hidden stale-artifact dependency.
  • If R2 surfaces sorryAx or an unexpected axiom in the closure, block acceptance and route to @lean-axiom-closure / @lean-proof.

Handoffs

  • Predecessors: @lean-proof (sorry removed), @lean-proof-review (single-proof review).
  • Successors: @lean-axiom-closure (R2 compile-time lockdown), @lean-quality-engine (milestone QA gate).

Detailed reference

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

See also

  • ../../references/lean-replay-verification-handbook.md -- Complete lean-replay-verification 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.

Check your setup, free