ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified developer tools Safety 4/5

openmath-open-theorem

Queries open formal verification theorems from the OpenMath platform. Use when the user asks for a list of open theorems, wants Lean or Rocq-specific theorems, needs full detail for a theorem ID, or wants to download a theorem and scaffold a local proof workspace.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/bennyzhe/openmath-open-theorem
Or

What This Skill Does

The openmath-open-theorem skill serves as the primary interface for the OpenMath ecosystem within OpenClaw. It enables developers and mathematicians to interact with a vast repository of formal verification theorems. By bridging the gap between local development environments and the OpenMath platform, this skill allows users to discover, inspect, and prepare formal proofs. It provides specialized tools for fetching theorem lists, viewing deep-dive technical details for specific proofs, and scaffolding local workspaces tailored for either Lean or Rocq formal verification languages.

Installation

To integrate this skill into your environment, use the OpenClaw command-line interface. Run the following command in your terminal: clawhub install openclaw/skills/skills/bennyzhe/openmath-open-theorem Once installed, the skill performs a mandatory first-run check via python3 scripts/check_openmath_env.py. This ensures that your local environment is correctly configured with a preferred_language (Lean or Rocq) and a secure storage scope for your configuration file, openmath-env.json. Setup must be completed successfully before any API-based discovery or download functionality becomes active.

Use Cases

This skill is ideal for researchers, computer scientists, and students working on formal proofs. Use it to:

  • Browse available open challenges on the OpenMath platform to find your next project.
  • Scaffold a new project workspace based on a specific theorem ID, complete with boilerplate structure.
  • Integrate with other OpenClaw skills to manage the full lifecycle of a proof, from initial fetch to final submission and reward claiming.
  • Perform status checks on existing proofs to verify if your work has been correctly submitted and recognized by the network.

Example Prompts

  • "List all available open Lean theorems that are currently marked as unverified."
  • "Get the full details for theorem ID 402, and please set up a local workspace for me to start proving it."
  • "I have finished the setup for my environment. Can you fetch the latest list of Rocq-based theorems for me to review?"

Tips & Limitations

The openmath-open-theorem skill requires a strict configuration workflow. Always verify that your openmath-env.json exists in either ./.openmath-skills/ or ~/.openmath-skills/. The skill is designed to prevent accidental data contamination; therefore, it will not auto-fan-out to multiple languages if you have a preferred_language set. If you need to switch between Lean and Rocq, ensure you update your configuration file accordingly. Note that this skill manages discovery and scaffolding; for the actual heavy lifting of proving, it relies on integration with openmath-lean-theorem, and for submitting your work, it requires openmath-submit-theorem to ensure cryptographic integrity of your proof submissions. Always ensure your environment is clean before starting a new theorem discovery session.

Metadata

Author@bennyzhe
Stars4473
Views0
Updated2026-05-01
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-bennyzhe-openmath-open-theorem": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#formal-verification#lean#rocq#mathematics#theorem-proving
Safety Score: 4/5

Flags: network-access, file-write, file-read, external-api, code-execution