Check suggested findings against the code and understand the account budget used by agentic features.
Coding & development · Documented · Documentation checked 2026-10-04
Editor assistance and agents, GitHub cloud agent, and code review have different execution contexts.
Plan, supported IDE, repository permission, AI credits, and organization policy determine access.
The worked example is illustrative and has not been tested in a signed-in product account.
What it does and how it connects
Check suggested findings against the code and understand the account budget used by agentic features.
Copilot code review proposes findings and changes for a selected diff. Review effort, surface support, and access depend on the documented plan and client. AI credits are a usage relationship to inspect, not a fixed promise that every review costs the same.
Walkthrough
Use the product named in the access panel. The steps below combine documented controls with the editorial example.
- Choose the intended diff or pull request on a supported review surface.
- Request Copilot review using the documented client controls.
- Inspect each finding and reproduce the stated trigger where practical.
- Apply justified fixes, request another review where useful, and retain human merge review.
Worked example
Review a sample navigation change for links that fail without JavaScript. A useful finding explains the click path and broken destination. Style preferences should not be presented as demonstrated defects.
Review this navigation diff for incorrect destinations, keyboard problems, and data regressions. For each concern, give the trigger and evidence. Distinguish a reproduced failure from a suggested improvement. Do not approve or merge the pull request.
Inspect and verify the result
Validate important findings against the actual route and code. Check whether the review missed edge cases you know matter. Inspect usage in the account’s current plan documentation rather than relying on old premium-request terminology.
Limits and troubleshooting
AI review can miss defects or raise false alarms. Available effort and automatic-review controls vary. No findings does not mean no risk; meaningful behavioral checks and a human reviewer remain necessary.
Data, permissions, and actions
Code and review context follow the product’s account and policy handling. Review extensions may have additional tools. Keep confidential issue or service data out unless explicitly needed and permitted.
Related reading
Completions, chat, and editor context
Distinguish a code suggestion from an agent that can edit files and run commands.
An editor fix with a checked handoff
Move from a checked suggestion to a scoped IDE agent change.
From a clear issue to a reviewed pull request
Create a cloud-agent change and a human-reviewed pull-request handoff.
Evaluations: does the system do the job?
Create representative tests for quality, tool behavior, refusal, escalation, and the effects of a change.
Sources and verification
Documentation checked 2026-10-04. Examples are illustrative unless described otherwise.
Copilot code review · GitHub
Plans for GitHub Copilot · GitHub