lean4-theorem-proving

lean4-theorem-proving is a skill for Claude Code, Codex from benchflow-ai/skillsbench. It costs 68 tokens per session (2,144 once invoked), scanned A, original, Apache-2.0.

A guide and toolkit for proving mathematical statements and verifying programs in Lean 4, a programming language and proof assistant. It covers incremental builds, compiler-guided fixes, mathlib lemma searches, and removing unfinished proofs.

In plain words
What is it for?
Use it to develop Lean 4 proofs, handle type-class errors, find mathlib lemmas, refactor proofs, and verify that no sorries or custom axioms remain.
Why use it?
It helps diagnose Lean errors and turn provisional proofs into checked results. Lean's type checker verifies whether the proof is correct.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents; mentions Claude Code.

Good fit Use it to develop Lean 4 proofs, handle type-class errors, find mathlib lemmas, refactor proofs, and verify that no sorries or custom axioms remain.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/benchflow-ai/skillsbench/lean4-theorem-proving
About the project

SkillsBench is a benchmark for measuring how effectively AI agents use modular skills—folders containing instructions, scripts, and resources—to complete specialized tasks. It helps researchers and developers evaluate both skill quality and agent behavior, including tasks that require combining multiple skills. The catalogue’s skills and instructions are evaluated as part of this workflow.

benchflow-ai/skillsbench · 1,764 stars · on GitHub · skillsbench.ai

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add benchflow-ai/skillsbench --skill lean4-theorem-proving
Clone the repo
git clone --depth 1 https://github.com/benchflow-ai/skillsbench

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for lean4-theorem-proving

README.md
[![agentmods](https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/lean4-theorem-proving/github.svg)](https://agentmods.dev/skills/benchflow-ai/skillsbench/lean4-theorem-proving)
Your own site
<a href="https://agentmods.dev/skills/benchflow-ai/skillsbench/lean4-theorem-proving"><img src="https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/lean4-theorem-proving/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for lean4-theorem-proving

Your own site · 80×15
<a href="https://agentmods.dev/skills/benchflow-ai/skillsbench/lean4-theorem-proving"><img src="https://agentmods.dev/badge/skills/benchflow-ai/skillsbench/lean4-theorem-proving.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 68 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,144 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium analysis-evasion · line 1
    Suspicious Unicode normalization or mixed-script content
    Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00068 $0.02144
Opus 5 $0.00034 $0.01072
Sonnet 5 $0.00014 $0.00429
Haiku 4.5 $0.00007 $0.00214

Measured 8d ago against content hash ccdf91ca8b6e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

lean4-theorem-proving scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

tasks/lean4-proof/environment/skills/lean4-theorem-proving/SKILL.md · 156 lines

How it starts

The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Lean 4 Theorem Proving

Core Principle

Build incrementally, structure before solving, trust the type checker. Lean's type checker is your test suite.

Success = lake build passes + zero sorries + zero custom axioms. Theorems with sorries/axioms are scaffolding, not results.

Quick Reference

Resource What You Get Where to Find
Interactive Commands 10 slash commands for search, analysis, optimization, repair Type /lean in Claude Code (full guide)
Automation Scripts 19 tools for search, verification, refactoring, repair Plugin scripts/ directory (scripts/README.md)
Subagents 4 specialized agents for batch tasks (optional) subagent-workflows.md
LSP Server 30x faster feedback with instant proof state (optional) lean-lsp-server.md
Reference Files 18 detailed guides (phrasebook, tactics, patterns, errors, repair, performance) List below

When to Use

Use for ANY Lean 4 development: pure/applied math, program verification, mathlib contributions.

Critical for: Type class synthesis errors, sorry/axiom management, mathlib search, measure theory/probability work.

Tools & Workflows

7 slash commands for search, analysis, and optimization - type /lean in Claude Code. See COMMANDS.md for full guide with examples and workflows.

16 automation scripts for search, verification, and refactoring. See scripts/README.md for complete documentation.

Lean LSP Server (optional) provides 30x faster feedback with instant proof state and parallel tactic testing. See lean-lsp-server.md for setup and workflows.

Subagent delegation (optional, Claude Code users) enables batch automation. See subagent-workflows.md for patterns.

Read the full file on GitHub · 156 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 156 lines · 68 tokens per session scan A ccdf91ca8b6e

Subscribe to this mod's changes

lean4-theorem-proving is a skill published in the GitHub repository benchflow-ai/skillsbench (1,764 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 68 tokens to every session and 2,144 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.