OpenMath Lean Theorem
name: openmath-lean-theorem
by bennyzhe · published 2026-04-01
$ claw add gh:bennyzhe/bennyzhe-openmath-lean-theorem---
name: openmath-lean-theorem
description: Configures Lean environments, installs external proof skills, runs preflight checks, and guides the workflow for proving downloaded OpenMath Lean theorems locally.
version: v1.0.2
requirements:
commands:
- python3
- lean
- lake
- elan
environment_variables:
- OPENMATH_SKILL_DIRS
- OPENMATH_LEAN_SKILL_INSTALL_DIR
- OPENMATH_LEAN_SKILL_REPO_URL
side_effects:
- May clone the reviewed leanprover/skills repo and copy missing Lean skills into an explicit install dir only when preflight is run with --auto-install-skills and --install-dir (or OPENMATH_LEAN_SKILL_INSTALL_DIR)
---
# OpenMath Lean Theorem
Instructions
Set up the Lean proving environment, validate toolchains, and prove downloaded OpenMath theorems locally. Assumes the theorem workspace was already created by the `openmath-open-theorem` skill.
This skill does not run benchmark providers, prompt-based agent comparisons, or trace persistence workflows. Those belong to the separate `openmath-lean-benchmark` skill.
Workflow checklist
```bash
npx leanprover-skills install lean-proof
npx leanprover-skills install mathlib-build
```
If you use preflight auto-install, first review the upstream repo and then pass an explicit target such as `--install-dir .codex/skills` or `--install-dir .claude/skills` so the write location is deliberate. Do not run auto-install without an explicit install dir.
Scripts
| Script | Command | Use when |
|--------|---------|----------|
| Preflight check | `python3 scripts/check_theorem_env.py <workspace>` | After download, before proving; validates toolchain, required skills, and initial build. |
| Preflight (auto) | `python3 scripts/check_theorem_env.py <workspace> --auto-install-skills --install-dir <path>` | Auto-install missing Lean skills during preflight into an explicit skills dir. |
Notes
References
Load when needed (one level from this file):
More tools from the same signal band
Order food/drinks (点餐) on an Android device paired as an OpenClaw node. Uses in-app menu and cart; add goods, view cart, submit order (demo, no real payment).
Sign plugins, rotate agent credentials without losing identity, and publicly attest to plugin behavior with verifiable claims and authenticated transfers.
The philosophical layer for AI agents. Maps behavior to Spinoza's 48 affects, calculates persistence scores, and generates geometric self-reports. Give your...