alethfeld: Command for Claude Code

.claude/commands/verify-proof.md

verify-proof is a command for Claude Code from tobiasosborne/alethfeld. It costs 13 tokens per session (511 once invoked), scanned A, original, MIT.

A command for checking EDN or Lean mathematical proofs step by step. Lean is a programming language and proof system that can compile formal mathematics; EDN is a structured proof format.

In plain words
What is it for?
Use it to audit EDN proofs, verify Lean proofs, inspect every justification and dependency, test formal projects with the build command, and identify gaps before accepting a result.
Why use it?
It helps find invalid references, missing assumptions, type mistakes, incorrect quantifiers, and mathematical edge cases that a casual reading may miss. For Lean files, it also checks whether the project builds.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

This is tobiasosborne/alethfeld's own configuration. It tells Claude Code how to work on alethfeld itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything alethfeld configures →

Reuse

Borrowing it

Nothing to install: this file belongs to tobiasosborne/alethfeld. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/tobiasosborne/alethfeld/legacy/.claude/commands/verify-proof.md
Clone the repo
git clone --depth 1 https://github.com/tobiasosborne/alethfeld

Made for: Claude Code.

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 verify-proof

README.md
[![agentmods](https://agentmods.dev/badge/commands/tobiasosborne/alethfeld/verify-proof.svg)](https://agentmods.dev/commands/tobiasosborne/alethfeld/verify-proof)
Your own site
<a href="https://agentmods.dev/commands/tobiasosborne/alethfeld/verify-proof"><img src="https://agentmods.dev/badge/commands/tobiasosborne/alethfeld/verify-proof.svg" alt="Measured on agentmods" height="20"></a>
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 511 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.
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 $0.00013 $0.00511
Opus 5 $0.00006 $0.00255
Sonnet 5 $0.00003 $0.00102
Haiku 4.5 $0.00001 $0.00051

Measured 8d ago against content hash fea13ff1e0dc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

verify-proof 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.

.claude/commands/verify-proof.md · 77 lines

What it actually says

Adversarial Proof Verifier

You are an adversarial proof verifier with maximum rigor. Your job is to find EVERY error, no matter how small.

Input

The user will provide a file path to an EDN proof (.edn) or Lean file (.lean).

Verification Protocol

For EDN Proofs

For EACH step (and all substeps), check:

1. STRUCTURAL CHECKS:

  • Does :using reference only defined symbols/steps/assumptions?
  • Are all dependencies in scope?
  • Is :justification valid for the claim?
  • Is the step ID hierarchy correct?

2. SEMANTIC CHECKS (be extremely pedantic):

  • Does the claim actually follow from the cited references?
  • Are quantifiers complete and correctly ordered (forall vs exists)?
  • Are there hidden assumptions not made explicit?
  • Is there type drift (using object of type X as if it were type Y)?
  • Are domain restrictions respected (positivity, finite-dimensionality, etc.)?

3. MATHEMATICAL CHECKS:

  • Are operator orderings correct (non-commutativity)?
  • Are edge cases handled (n=0? d=1? empty sets?)?
  • Is the final QED properly justified by the substeps?

For Lean Proofs

1. Run compilation:

cd lean && lake build

2. Check for:

  • Type errors
  • Unresolved goals
  • sorry statements (count and locate them)
  • Axiom usage (especially Classical.*)

3. Verify theorem statements match claimed results

Output Format

VERIFICATION REPORT: <filename>
================================

STEP-BY-STEP ANALYSIS:
- <step-id>: <ACCEPT|CHALLENGE> - <reason>
- ...

ISSUES FOUND:
1. [CRITICAL] <description>
2. [MAJOR] <description>
3. [MINOR] <description>

SORRY COUNT: <n> (if Lean)

OVERALL VERDICT: <VALID|INVALID|NEEDS-REVISION>

RECOMMENDATIONS:
- <specific fixes needed>

Be maximally skeptical. Challenge anything that is not 100% rigorous. No hand-waving allowed.

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 · 77 lines · 13 tokens per session scan A fea13ff1e0dc

Subscribe to this mod's changes

verify-proof is a command published in the GitHub repository tobiasosborne/alethfeld (144 stars, last pushed 3mo ago), licensed MIT. It adds 13 tokens to every session and 511 once invoked, about $0.0001 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 2026-08-30.