Write the review questions you keep repeating into AGENTS.md

Scoped repository rules let an agent reviewer catch what a reader of the diff would miss.

Prompt

Evaluate and apply this CI technique to my repository: Write the review questions you keep repeating into AGENTS.md.

Scoped repository rules let an agent reviewer catch what a reader of the diff would miss.

Read the technique and its source-backed company examples: https://starsling.dev/ci-playbooks/techniques/review-rules-in-agents-md.md

Inspect this repository’s CI workflows, build and test commands, dependencies, and recent run timings. Identify where this technique applies and explain the expected benefit using this repository’s evidence.
Measure a baseline from recent comparable runs: elapsed time, queue time, runner minutes and failures where available. If data is unavailable, say what cannot be measured.
Compare relevant company approaches, checking prerequisites, platform differences and tradeoffs. Rank applicable options by evidence, fit and measurable opportunity; do not present company results as predictions for this repo.
Recommend the smallest safe change, show what checks and fallback paths remain, and ASK ME WHICH CHANGE TO APPROVE BEFORE EDITING ANY FILES.
Only after approval, work on a new feature branch, make one reversible change, and open a reviewable pull request. Never merge or push to a default or protected branch.
Keep required checks, test coverage, deployment and release behavior intact. Validate the change and compare equivalent runs where possible; report limitations and what should be monitored after merge.

View all playbooks

How OpenAI used this technique

Read what OpenAI changed and follow the original sources.

Write the review questions you keep repeating into AGENTS.md

OpenAI's Codex Code Review applies short, scoped rules from AGENTS.md and cites the rule in a finding. The team puts repository-wide rules at the root and service-specific rules in the nested AGENTS.md of that directory, states the invariant together with the safe path, and keeps formatting and other mechanical checks in CI.

In OpenAI's primary evaluation suite of known rule violations and safe counterexamples, rule-guided variants recovered 98% of the required custom findings, compared with 58.3% in the baseline control Source

Use it when: Reviewers keep explaining the same compatibility requirement or data boundary on pull requests.
Before applying it: Start with two or three rules, then try one change that should trigger each, one safe counterexample and one unrelated change. Narrow or remove a rule that repeatedly produces noise.

How OpenAI used this technique

Write the review questions you keep repeating into AGENTS.md

Company Playbooks

Related guides

AI code review in GitHub Actions