Review an AI-assisted code change before release
Move an AI-assisted code change from a clear requirement through security, testing and accountable human approval.
What you will produce
A reviewable result, not an automatic answer.
Follow the stages in order. Each stage ends with a tangible deliverable and a review checkpoint, so mistakes are caught before they become the next tool’s input.
Use free plans where suitable; confirm current provider limits and prices before starting. Compare the latest verified AI pricing before buying another subscription.
Suggested stack
Tools used in this workflow.
The process matters more than the brand. Use an approved equivalent when it offers the controls and output quality your organisation needs.
Inputs and ownership
Before you start
- A version-controlled codebase, issue and acceptance criteria
- A safe development environment without production secrets
- An authorised reviewer who understands the affected system
Interactive workflow
Complete each stage in order.
Tick a stage only after its deliverable and review checkpoint are complete.
-
Translate the requirement into acceptance criteria, affected components, constraints and a rollback expectation.
DeliverableA reviewable engineering brief
Review checkpointNever expose credentials, customer data or production configuration.
-
Ask the assistant to inspect relevant code and propose a small change plan before generating code.
DeliverableA bounded implementation plan
Review checkpointInspect every changed line and dependency.
-
Implement the smallest coherent change with comments only where they explain non-obvious decisions.
DeliverableA focused code change
Review checkpointTest failure paths, permissions and untrusted input.
-
Run formatting, static analysis, unit, integration, security and regression checks appropriate to the risk.
DeliverableA recorded verification result
Review checkpointReject unrelated refactors hidden inside the change.
-
Complete human code review, deployment review, monitoring and rollback preparation before release.
DeliverableAn approved release and rollback plan
Review checkpointMonitor the deployed behaviour against the acceptance criteria.
Adapt, do not paste blindly
Working prompts for this recipe.
Replace bracketed placeholders, supply approved sources and remove unnecessary personal or confidential information.
Working prompt
Use this only after replacing every bracketed placeholder and removing unnecessary sensitive information.
Review the code task below and the supplied repository context. First restate acceptance criteria, identify affected components, security boundaries, tests and unknowns. Then propose the smallest change plan. Do not write code until missing information is listed. Preserve existing behaviour outside scope and never request or expose secrets.\n\nTASK:\n[TASK]\nCONTEXT:\n[APPROVED CODE CONTEXT]
Final review prompt
Use as a second-pass checklist, not as approval to publish.
Review the draft against this intended outcome: Move an AI-assisted code change from a clear requirement through security, testing and accountable human approval. List unsupported claims, missing evidence, ambiguous language, privacy concerns, accessibility issues and actions that still require human approval. Do not rewrite the draft until the issues are listed separately.
Human review required
Verification and cautions
- Verify names, figures, quotations and source claims before use.
- Remove confidential or personal information unless the selected service and account are approved for it.
- Keep a named person responsible for the final decision and published output.
Keep the outcome, change the method
Practical alternatives
Continue with context