メニュー
LLM-assisted tools for informal proofs, proof strategy discussion, and code simplification
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
A no-op skill used only by the agent-capability tests to confirm a harness can mount and invoke a skill. Has no real behavior; invoke it when asked to during a capability probe.
Use when editing .lean files, debugging Lean 4 builds (type mismatch, sorry, failed to synthesize instance, axiom warnings, lake build errors), searching mathlib for lemmas, formalizing mathematics in Lean, or learning Lean 4 concepts. Also trigger when the user asks for help with Lean 4, mathlib, or lakefile. Do NOT trigger for Coq/Rocq, Agda, Isabelle, HOL4, Mizar, Idris, Megalodon, or other non-Lean theorem provers.
Lean 4 theorem proving toolkit: search lemmas, verify proofs, repair/simplify code, and get LLM-assisted informal proofs
Code transformation tools for repairing, simplifying, and extracting Lean proofs
Search tools for finding Lean theorems, lemmas, and definitions in Mathlib
Isolates failing proof steps by replacing them with `sorry` and extracting them into standalone lemmas to modularize and decouple complex Lean 4 proofs.
| name | llm |
| description | LLM-assisted tools for informal proofs, proof strategy discussion, and code simplification |
LLM-assisted tools for theorem proving support. All scripts are in .claude/skills/cli/.
| Tool | Purpose | When to use |
|---|---|---|
| informal-prover | Generate and verify step-by-step math solutions in a loop | When you want an LLM to attempt a full solution with auto-verification |
| discussion-partner | Free-form discussion about proof strategies or Lean code | When you are stuck and want high-level strategic advice |
| code-golf | Shorten and simplify an existing Lean proof via Gemini | After a proof works, to get a more elegant version |
For full parameters and examples, read the corresponding reference-<tool>.md file in this directory.
