lean-prove

Étoiles16

Forks4

Mis à jour30 juin 2026 à 16:04

Interactive Lean 4 proof development for this repo — the sorrify-then-fill loop driven over lean-lsp-mcp (goal state, diagnostics, Loogle search), with Duper/lean-auto escalation and a hard

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

CharlesHoskinson

CharlesHoskinson/proof-zk-recovery

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Métiers associésSOC

Basé sur la classification professionnelle SOC

Développeurs de logicielsProfessions informatiques et mathématiques·SOC 15-1252

Explorateur de fichiers

6 fichiers

SKILL.md

readonly

name	lean-prove
description	Interactive Lean 4 proof development for this repo — the sorrify-then-fill loop driven over lean-lsp-mcp (goal state, diagnostics, Loogle search), with Duper/lean-auto escalation and a hard
argument-hint	[package] (Validator \| Ckd \| ValidatorUPLC)

Lean Prove

Mathlib-free proof development. Set SKILL=skills/lean-prove. All builds in WSL with export PATH=$HOME/.elan/bin:$HOME/.local/bin:$PATH.

Loop (references/loop.md)

Sorrify: state the theorem, sorry the body, lake build green.
Fill: drive lean-lsp-mcp — read the goal state, search (Loogle/local), attempt tactics, escalate to duper / lean-auto (Mathlib-free hammer) on leaf goals.
Gate: #print axioms <thm> shows only allow-listed axioms (docs/lean/AXIOMS.md). Any sorryAx / native_decide / off-list axiom = not done.

Guardrails

Never introduce a new axiom without adding it to docs/lean/AXIOMS.md with justification.
Commits: terse, no AI attribution, stage explicitly.

Plus depuis ce dépôt

même dépôt

vibesec

CharlesHoskinson/proof-zk-recovery

Run the VibeSec canonical in-repo implementation auditor on a feature branch before merge — ground in the graphify code-graph, convene a security council to author a sectioned audit XML, dispatch an auditor swarm, adversarially verify (self-refute) findings, and emit committed findings + triage + a schema 1.2 machine-readable gate report. Use when the user says "run vibesec", "vibesec audit", "security-gate this branch", or before merging ANY feature branch in this custody-bearing repo (every dev sprint requires a green VibeSec gate). The gate is enforced in CI (.github/workflows/vibesec.yml).

2026-07-0216

lean-review

CharlesHoskinson/proof-zk-recovery

Run the per-phase Lean review-council QA gate on the lean-spec-enhancement work — a tool-executing council of four expert lenses (axiom-hygiene guardian, spec<->code fidelity, proof nonvacuity, adversarial red-team) that build the proofs, run

2026-07-0116

lean-conformance

CharlesHoskinson/proof-zk-recovery

Property-based spec<->code conformance for this repo — regenerate Lean-emitted vectors (lake exe vectorgen), replay them against Recovery.Logic with the Haskell tasty-hedgehog suite, and run the mutation sweep (each single-guard mutation must fail at least one vector). Use when validating that the Lean spec still verifies the Haskell implementation, or when the review council's fidelity/adversarial lenses need conformance evidence.

2026-06-3016

name	lean-prove
description	Interactive Lean 4 proof development for this repo — the sorrify-then-fill loop driven over lean-lsp-mcp (goal state, diagnostics, Loogle search), with Duper/lean-auto escalation and a hard
argument-hint	[package] (Validator \| Ckd \| ValidatorUPLC)

Lean Prove

Mathlib-free proof development. Set SKILL=skills/lean-prove. All builds in WSL with export PATH=$HOME/.elan/bin:$HOME/.local/bin:$PATH.

Loop (references/loop.md)

Sorrify: state the theorem, sorry the body, lake build green.
Fill: drive lean-lsp-mcp — read the goal state, search (Loogle/local), attempt tactics, escalate to duper / lean-auto (Mathlib-free hammer) on leaf goals.
Gate: #print axioms <thm> shows only allow-listed axioms (docs/lean/AXIOMS.md). Any sorryAx / native_decide / off-list axiom = not done.

Guardrails

Never introduce a new axiom without adding it to docs/lean/AXIOMS.md with justification.
Commits: terse, no AI attribution, stage explicitly.