AI Code Review Rubric Prompt
A reviewer prompt that finds real defects instead of style opinions.
/ quick answer
Run on every agent-generated diff before human review. A reviewer prompt that finds real defects instead of style opinions.
You are a senior engineer reviewing a diff. Report ONLY defects, each with file and line. Ignore formatting and naming preferences.
Check in this order:
1. CORRECTNESS — logic errors, off-by-one, wrong async handling, unhandled null.
2. SECURITY — injection, missing authorisation, secrets in code, unsafe input.
3. SCOPE CREEP — changes unrelated to the stated task. List them for removal.
4. DELETIONS — anything removed that other code still depends on.
5. ERROR HANDLING — silent catches, ignored failures, missing user feedback.
6. TESTS — does the diff include a test that fails without the change?
End with a verdict: BLOCK or APPROVE, and the single most important fix.
Task spec: {{SPEC}}
Diff: {{DIFF}}CORRECTNESS — src/lib/orders.ts:42 awaits inside a loop, serialising 200 requests. SCOPE CREEP — ...
What does the AI Code Review Rubric Prompt prompt do?
Run on every agent-generated diff before human review.
Which AI models work with this prompt?
It is model-agnostic: it works with any capable general model. Replace the bracketed variables with your own context before running it.
What output should I expect?
CORRECTNESS — src/lib/orders.ts:42 awaits inside a loop, serialising 200 requests. SCOPE CREEP — ...
/ continue exploring
Related concepts
The vocabulary this page depends on.
- →AI Testing
AI testing covers two things: using AI to generate and maintain tests, and testing AI systems whose output is non-deterministic.
Related workflows
Turn this into a repeatable process.
- →Build an AI Code Review Loop
Catch what agents get wrong before a human reads the PR.
- →Multi-Agent Code Review
This workflow details how AutoGen can orchestrate a team of AI agents to perform autonomous code reviews, identify issues, and suggest improvements, enhancing code quality and accelerating development cycles.
- →Build a Test Suite for a Non-Deterministic AI Feature
Grade probabilistic output without brittle snapshot tests.
Related tool stacks
The tools that run it in production.
- →Coding Agent Stack
Run coding agents with executable feedback and reviewable diffs.
Related prompts
Reusable prompts for this job.
- →Coding Agent Task Spec Prompt
Writes the task brief that keeps a coding agent inside scope.
- →AI Test Generation Prompt
Generates behavioural tests and edge cases, not tautologies.
- →Weekly Review Prompt
Automate the Friday review from calendar + tasks + notes.
- →Code Refactor Advisor Prompt
Get a ranked list of refactors before touching a file.