This table is the evidence-based summary of bots and checks on pull requests. For branch rules and CI layout, see .github/INFRASTRUCTURE.md and ci.yml.
| Bot / service | Role | Enforcement | Repo evidence |
|---|---|---|---|
| Claude Review | PR review + assist flows | Required check on protected branches | .github/workflows/claude-review.yml, Claude Review check |
| GitHub Copilot code review | Automatic PR review | GitHub rulesets with copilot_code_review on develop, main, release/*, and hotfix/* |
Branch rulesets + .github/copilot-instructions.md |
| Seer (Sentry) | Safety / error-handling review comments | Required check plus unresolved AI-thread gate | .github/seer.app.yml, Seer Code Review, Autonomous AI Review |
| SonarQube Cloud | Quality gate, PR decoration | Required check on protected branches | SonarCloud Code Analysis, config/sonar-project.properties |
| Cursor BugBot | Bug-first review policy / repo guidance | Policy contract in repo; not a live GitHub status check unless the org-level app is installed separately | BUGBOT.md |
Protected-branch baseline:
develop,main,release/*, andhotfix/*require GitHub Copilot code review.developand release branches require the core CI gates plus Claude, Seer, Sonar, and platform build/test checks before merge.- Unresolved AI review threads from Copilot and Sentry fail
Autonomous AI Review.
Update this doc when tooling changes; do not duplicate long benchmark tables here — see CLAUDE.md for North Star and budget context.