Lean doc improvement
APM-installable agent skills for Lean 4 and Mathlib4 — proof tactics, math domains, review and research workflows, generic tooling.
npx -y skills add r-irbe/proof-skills --skill lean-doc-improvementAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
USE FOR: Update academic papers, technical reports, and documentation based on results from Lean 4 formalization. Use when formal verification reveals paper imprecisions, missing hypotheses, sharper bounds, new insights, or when metrics need synchronization. Covers paper appendix updates, verification tables, erratum entries, insight propagation, and metric synchronization between Lean and documents. DO NOT USE FOR: extracting requirements from papers (use @lean-doc-requirements); blueprint generation (use @lean-blueprint); report compilation (use @lean-report). TRIGGERS: update paper, doc improvement, paper revision, documentation update from Lean, doc-feedback consumer.
SKILL.md
3.7 KB, as published. Nobody here has run it
Lean 4 Results-to-Document Improvement
Systematically update documents when Lean formalization produces new insights, corrections, or metrics.
Routing
- USE FOR: Update academic papers, technical reports, and documentation based on results from Lean 4 formalization. Use when formal verification reveals paper imprecisions, missing hypotheses, sharper bounds, new insights, or when metrics need synchronization. Covers paper appendix updates, verification tables, erratum entries, insight propagation, and metric synchronization between Lean and documents.
- DO NOT USE FOR: extracting requirements from papers (use @lean-doc-requirements); blueprint generation (use @lean-blueprint); report compilation (use @lean-report).
- TRIGGERS: update paper, doc improvement, paper revision, documentation update from Lean, doc-feedback consumer.
Workflow
- Triage the change: which paper/document needs updating, which Lean result triggers it, and what's the delta.
- Pick the section/template from the body matching the update class (definition, theorem, proof-strategy, errata).
- Produce the patch with precise pointers to the Lean source + before/after text.
- Hand off: to
@lean-doc-feedbackfor the doc-sync record, to@lean-reportif the paper feeds into a report, to@lean-zettelkasten.
Recovery & STOP
- STOP if the question is about extracting requirements from the paper — delegate to
@lean-doc-requirements. - STOP if a fresh blueprint is needed — delegate to
@lean-blueprint. - STOP if the Lean result is wrong (not the paper) — delegate to
@lean-proof-reviewfirst.
Handoffs
- Predecessors:
agent:gateway,skill:lean-research. - Successors:
skill:lean-doc-feedback,skill:lean-report,skill:lean-zettelkasten.
Detailed reference
Full content for lean-doc-improvement lives in
references/lean-doc-improvement-handbook.md.
Load that file when the skill is convened; the SKILL.md only carries
the dispatch contract and the parts index.
| Section | Topic |
|---|---|
| Part 1 | Trigger Events |
| Part 2 | Document Update Types |
| Part 3 | Update Workflow |
| Part 4 | LaTeX Integration Patterns |
| Part 5 | Feedback to Other Skills |
| Part 6 | Metric Dashboard |
See also
../../references/lean-doc-improvement-handbook.md— Full handbook (extracted from this skill)../lean-doc-feedback/SKILL.md— Successor../lean-report/SKILL.md— Successor../lean-zettelkasten/SKILL.md— Successor