proof-agent

Community

Block unsafe AI changes with evidence.

AuthorAndreaGriffiths11
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Proof Agent prevents AI-generated code changes from slipping through review by requiring an independent verifier to check security, correctness, and factual claims using concrete evidence rather than self-approval.

Core Features & Use Cases

  • Adversarial, separate-agent verification: Spawns an independent verifier so the worker cannot validate its own output.
  • Static diff-based security review: Reviews the git diff without executing code or running tests, and requires specific file/line/snippet evidence for verdicts.
  • Auto-verify triggers for high-risk edits: Verifies automatically when enough files change or when sensitive patterns appear (auth/secret/permission, Dockerfile, .env).

Quick Start

Tell your agent: “Verify that this change is safe for production and report PASS, FAIL, or PARTIAL with file/line evidence.”

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: proof-agent
Download link: https://github.com/AndreaGriffiths11/proof-agent/archive/main.zip#proof-agent

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.