Agents PR Review Evidence Discipline — AI skill for Claude Code
Four codified rules for reviewing AI-agent authored PRs that drop a project prefix or rename a package.
How to install Agents PR Review Evidence Discipline
Installs to ~/.claude/skills/homericintelligence-projectmnemosyne-agents-pr-review-evidence-discipline/SKILL.md
mkdir -p ~/.claude/skills/homericintelligence-projectmnemosyne-agents-pr-review-evidence-discipline && curl -fsSL https://raw.githubusercontent.com/HomericIntelligence/ProjectMnemosyne/HEAD/skills/agents-pr-review-evidence-discipline.md -o ~/.claude/skills/homericintelligence-projectmnemosyne-agents-pr-review-evidence-discipline/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Agents PR Review Evidence Discipline does
name: agents-pr-review-evidence-discipline license: BSD-3-Clause description: "Four codified rules for reviewing AI-agent authored PRs that drop a project prefix or rename a package. Use when reviewing chore/rename-* PRs, when claiming a PR is CLEAN/DIRTY, or when posting a verdict comment. Prevents false-positive NO-GO verdicts caused by inspecting the wrong branch." category: tooling date: 2026-07-11 version: "1.0.0" user-invocable: false verification: verified-local tags: [pr-review, evid
Alternatives in AI
- Debate — AI Debate Hub - Structured four-way debates between Claude, Sonnet, Gemini, and Codex 2.8k ★
- Meme Rush — Track real-time meme token lists from launchpads (Pump.fun, Four.meme) and AI-powered hot market topics ranked 483 ★
- OpenAIReview — AI reviewing paper drafts for improvement 160 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Rtk Commit Message Style
Conventional Commits enforcer for AI-authored PRs.
Garl Receipt Action
Cryptographic verification for AI-authored code. ECDSA-signed receipts (secp256k1, RFC 6979 deterministic) for
B2B AI FDE Contract Counsel
Four-lens contract counsel for AI B2B deals — legal, regulatory, finance/tax, commercial. A Claude Code skill
Leo
A deterministic context-driven agentic SDLC framework for Cursor and Claude. 41 codified laws and 22 specialis
Agent Roadmap
Quick-and-dirty release roadmap for working with AI coding agents (Claude Code, Codex, Cursor). One HTML file
Canny
Stops AI coding agents from claiming work is done without evidence. Deterministic hooks decide, TypeSafe's Jev
Related Agents
AI PR Gate
Reviews bot-authored PRs for INV-91 compliance. Checks that the approved-by-human label is present and that th
Strict Skill Creator
Use when the user wants a new skill authored, or an existing skill reworked, in the strict.ai repository — wri
Design Composition Critic
Specialist critic · focuses ONLY on composition — section order, grid proportions, whitespace rhythm, hollow-c