They already run local reviews with Codex and Claude Code. Assigned as a reviewer in GitHub, Copilot finishes in a couple of minutes and raises valid issues they'd never found.
Result Review done within a couple of minutes
“We just assign it as a reviewer in GitHub and a couple minutes later, the review is done.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
Besides running local code review with Codex and Claude Code, they assign GitHub Copilot with custom instructions as a reviewer directly in GitHub on their pull requests.
A couple of minutes after being assigned, Copilot’s review is done. It raises a lot of issues that are valid and that they had never found themselves.
Configure an AI code review tool like GitHub Copilot with custom review instructions for your repository's conventions. Assign it as a reviewer on your pull requests the same way you would a human teammate, alongside any local AI code review you already run. Treat every issue it raises as a real lead to check, not noise.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.