pr-rulebook
Compile your team's implicit code-review rules from accepted GitHub PR feedback into a rulebook for Cursor, Claude Code and CodeRabbit. Local-first, evidence included.
by ofershapTypeScript
Meet the team
See all 1 on GitHub →
Commit activity
Last 12 weeks · 40 commits
Community health
2 of 6 standards met
Recent PRs & issues
PR Rulebook compiles recurring human review feedback into evidence-backed rules for Cursor, Claude Code, and CodeRabbit. The first real run on astral-sh/ruff found one coherent candidate and one bad cluster. That failure led to two changes: candidate rules must now span multiple PRs, and clustering ignores fenced suggestions and normalizes related review terms. I am looking for 5 public repositories with active human PR review. You will get the candidate rules and evidence links back, with false positives reported honestly. The scanner runs locally - no code or comments go to a hosted service. To volunteer, comment with: repository URL rough number of merged PRs per month any paths to exclude Real experiment: https://github.com/ofershap/pr-rulebook#real-repository-experiment Visual project page: https://gitshow.dev/ofershap/pr-rulebook
Repository: ofershap/pr-rulebook. Description: Compile your team's implicit code-review rules from accepted GitHub PR feedback into a rulebook for Cursor, Claude Code and CodeRabbit. Local-first, evidence included. Stars: 0, Forks: 0. Primary language: TypeScript. Languages: TypeScript (100%). License: MIT. Homepage: התגיות: code-review, cursor, claude-code, coderabbit, ai-coding, developer-tools, typescript, llm, cli, context-engineering, agents-md, pr-feedback ובשדה Website: https://files.instinct.com/file-01M2XTYEPCK4SXC9JBC2891V68 Topics: agents-md, agentsmd, ai-coding, claude, claude-code, cli, coderabbit, coderabbitai, context-engineering, core-review, cursor, developer-tools, llm, pr-feedback, typescript. Open PRs: 0, open issues: 1. Last activity: 1mo ago. Community health: 42%. Top contributors: ofershap.