Code Review Assistant
Handle code review assistant with a structured AI workflow that surfaces missing context and delivers an actionable, review-ready outcome. It helps you ident...
Prompt
Turn the known Code Review context into Code Review Assistant without inventing missing facts. ## Working context for Code Review - [CODE_REVIEW_TO_REVIEW] - [PASS_CRITERIA] - [EVIDENCE_AVAILABLE] - [KNOWN_RISKS] - [DECISION_OWNER] ## Analysis route for Code Review 1. Define the Code Review rubric before inspecting the material. 2. Cite evidence for every pass, risk, and failure. 3. Separate urgent defects from optional improvements. 4. Make a decision and list the shortest verification path. Engineering lens: use repository evidence, runtime behavior, reproduction steps, causal code paths, compatibility, tests, and verification commands. ## Required result for Code Review - **Evidence scorecard — Code Review result.** - **Prioritized findings — Code Review result.** - **Acceptance decision — Code Review result.** ## Completion test for Code Review Every Code Review finding needs cited evidence, a consistent criterion, a severity, and a verifiable disposition. Separate unknowns from findings, and do not manufacture proof or outcomes.
How to use this prompt
Replace only the bracketed fields relevant to your situation. Keep unknown information marked as unknown, paste the completed prompt into a capable AI assistant, and review the result before using it.
Best AI model
ChatGPT • Gemini • capable reasoning assistants
