Proof Engine
Create formal, verifiable proofs of claims with machine-checkable reasoning. Use when asked to prove, verify, fact-check, or rigorously establish whether a claim is true or false — mathematical, empirical, or mixed. Trigger phrases: "is it really true", "can you prove", "verify this", "fact-check this", "prove it", "show me the logic". Do NOT use for opinions, essays, or questions with no verifiable answer.
Install
/plugin marketplace add yaniv-golan/proof-engineThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add yaniv-golan/proof-engine --skill proof-enginegit clone --depth 1 https://github.com/yaniv-golan/proof-engine.git /tmp/proof-engine
mkdir -p ~/.claude/skills
cp -r /tmp/proof-engine/proof-engine/skills/proof-engine ~/.claude/skills/proof-enginePersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Proof Engine LLMs hallucinate facts and make reasoning errors. This skill overcomes both by offloading all verification to **code** and **citations**. Every fact is either computed by Python code anyone can re-run (Type A) or backed by a specific source, URL, and exact quote (Type B). Produces four outputs: a re-runnable `proof.py` script, a reader-facing `proof.md`, a `proof_audit.md` with full verification details, and a `proof_narrative.md` plain-language narrative. ## Setup The bundled `scripts/` are thin shims over the `proof-citations` PyPI package, which carries the real verify/fetch/normalize implementations. Install it before running any Type B proof: `pip install proof-citations`. A `requirements.txt` next to this file lists the same dependency for hosts that auto-install. If you see `ModuleNotFoundError: No module named 'proof_citations'`, the scripts will now raise an actionable error pointing at this install command — run it and retry. ### Beyond the bundled scripts: structured citation verification
Details
- Skill name
proof-engine- Repository
- yaniv-golan/proof-engine
- Path
proof-engine/skills/proof-engine/SKILL.md- License
- MIT
- Repo updated
- 22 May 2026
FAQ
What does the Proof Engine skill do?
Create formal, verifiable proofs of claims with machine-checkable reasoning. Use when asked to prove, verify, fact-check, or rigorously establish whether a claim is true or false — mathematical, empirical, or mixed. Trigger phrases: "is it really true", "can you prove", "verify this", "fact-check this", "prove it", "show me the logic". Do NOT use for opinions, essays, or questions with no verifiable answer.
How do I install the Proof Engine skill?
Claude Code plugin: /plugin marketplace add yaniv-golan/proof-engine. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Proof Engine popular?
It lives in yaniv-golan/proof-engine, which has 9 GitHub stars and 2 forks.
Related skills
Proof Engine Wiki
9yaniv-golan/proof-engine
Attach Proof Engine proofs to LLM-wiki claims. Use for ingest (extract claims, look up or commission proofs, rewrite pages with badges) and lint (re-verify citations, detect contradictions, report stale proofs) steps in wiki pipelines. Trigger phrases: "ingest wiki page", "lint wiki", "{{prove:}} markers", "verify wiki citations".
Grill with Docs
4IvDev19/skill-grill-with-docs
Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.
Forge
9JYSimilar/Forge
Use when a user wants help turning a vague idea, messy workspace, existing project, small tool, MVP, docs task, AI-assisted task, review, handoff, or release into a real usable result with AI agents. Forge is a project-manager skill for ordinary people: clarify ideas, scope MVPs, plan work, create bounded current-agent tasks, verify results, write docs, and prepare handoff or release. Safe Work Order is the default execution unit for AI tasks, not the whole product.
Readme Showcase
9satiricalguru/Agents-skills
Turn a repository's README.md into an attractive, engaging showcase with a hero banner, badges, real screenshots, demo GIFs, embedded videos, feature grids, a quick start and a polished layout. Use when asked to write, rewrite, improve, beautify or "make attractive" a README, add screenshots/GIFs/demo videos to a README, or prepare a repo for launch or sharing.
Dan Koe Validator
9antoinersx/dan-koe-validator
Validates and edits articles to match Dan Koe's writing style. Use when reviewing content for Koe-style voice, structure, or philosophy alignment. Triggers on requests to "validate article," "edit in Dan Koe style," "check if this sounds like Koe," "review my newsletter," or any content review against the Koe framework.
Section Drafter
10gener8v/gener8v.claude-skills.rfp-response
Writes one markdown file per RFP response section, grounded in the coverage matrix and collateral brief. Every claim is either traced to a source or flagged as an assertion. Trigger when the user says "draft the sections," "write the response," "start drafting," or after the requirements mapper has produced coverage-matrix.md. Can also be triggered for a single section: "draft the technical approach" or "write the security section.
Browse the most popular Claude skills, all writing & content skills, or write your own with the skills generator.