Integration
Use Juror in ChatGPT and Codex.
The Juror Plugin connects to hosted Juror Cloud through OAuth. It is not a local-repository agent: it exposes only workspace-scoped review metadata and does not return raw diffs, source checkouts, reports, artifacts, screenshots, credentials, or prompt text.
Direct answer
The Juror Plugin connects to hosted Juror Cloud through OAuth. It is not a local-repository agent: it exposes only workspace-scoped review metadata and does not return raw diffs, source checkouts, reports, artifacts, screenshots, credentials, or prompt text.
01Choose a workspace
02Inspect minimal review summaries
03Confirm hosted review starts
Supported review workflow
List your workspaces first. Inspect concise findings and hosted runs with a selected workspace ID. To run or rerun a review, Juror first preflights the open PR, current SHA, repository policy, GitHub Action conflict, and capacity. Show that result to the user and obtain an explicit confirmation before the write tool can start a hosted review.
Clear data boundaries
The integration returns minimal metadata and an explicitly requested retained finding body or claim. It never returns raw diffs, source checkouts, full reports, screenshots, artifacts, provider credentials, or prompt text. It runs hosted Juror Cloud—not an arbitrary repository on the agent machine.
Starter prompts
- Show my Juror workspaces and review findings.
- Summarize open P0 and P1 Juror findings in this workspace.
- Prepare a hosted review for this PR and show the SHA and capacity warning before starting.
How it works
The Juror Plugin connects to hosted Juror Cloud through OAuth. It is not a local-repository agent: it exposes only workspace-scoped review metadata and does not return raw diffs, source checkouts, reports, artifacts, screenshots, credentials, or prompt text.
For use juror in chatgpt and codex., use the published configuration and source repository as the product record. Keep the workflow small enough to inspect, and record any exception in the pull request rather than assuming a model result is final.
Limits to keep in view
Models, providers, and benchmark conditions change. Juror does not replace code ownership, test suites, static analysis, or a human decision to merge. Treat unknown cost and unevaluated compatibility as explicit unknowns.