AAWEA.ORG

Manual start

↓

1

Review diff

Review the current branch diff. List bugs, edge cases, naming issues, and missing tests.
2

Address findings

Fix the highest-severity findings from the review with minimal scope.
3

Re-review

Review the updated diff and confirm prior findings are resolved.

↓

✓

Exit condition: max_iterations

Three review passes completed with no critical findings

📊 Flow Diagram

flowchart TD Start(["Manual start"]) Start --> S0 S0("Review diff") S1("Address findings") S0 --> S1 S2("Re-review") S1 --> S2 Exit{"Three review passes completed with no critical findings"} S2 --> Exit Exit -- "No" --> S0 Exit -- "Yes" --> Done(["Success"]) style Exit fill:#fffbeb,stroke:#f59e0b,stroke-width:2px,color:#92400e style Done fill:#ecfdf5,stroke:#10b981,stroke-width:2px,color:#065f46

Run "PR Self-Review" in your agent

Prompt only

Deeplinks and "Open in Cursor" only paste the kickoff prompt. They do not install hook files — your agent cannot tell whether files are on disk until you add them yourself.

Two separate pieces

Kickoff prompt — tells the agent the goal, check command, exit condition, and how to self-pace.

Kickoff prompt

Start the "PR Self-Review" loop.

Goal: three clean self-review passes on the current diff
Max iterations: 3
Between iterations run: git diff main...HEAD
Exit when: three passes complete with no critical findings

Step 1: Review the diff like a senior reviewer. Fix findings, then re-review.

Self-pace this loop. After each iteration, run the check command, read the output, and only continue if the exit condition is not met. Stop when the exit condition passes or max iterations is reached. Give a short status update each pass.

Anti-gaming rules

Rules the agent must follow so it cannot cheat the exit condition.

Do not modify the check command or exit criteria to force success
Do not skip, disable, or bypass checks to pass the exit condition
If stuck after several iterations, stop and report blockers instead of gaming metrics