Turn a bug report into a reproduction plan
A clear diagnostic checklist that helps a developer reproduce the issue.
A little inspiration is on its way…
OpenAI's own headline example: a dot that watches feedback and issues for repeat requests, scopes the smallest fix, builds and tests it in Codex, and opens a PR with a recording for you to review.
Community-reported results. Success rates appear after 10 approved reports. Copies are deduplicated per visitor each day. Starter workflows have not been independently tested.
Small fixes pile up behind the big feature. A dot with a fork and a test environment can clear them one PR at a time while you keep building.
Source / inspiration: View credited source
Help the next person know what to expect.
Sign in to share your resultA question, a useful tweak, a little discovery. Start the conversation.
Sign in to join inA clear diagnostic checklist that helps a developer reproduce the issue.
Paste this into a conversation with your dot and replace anything in [square brackets] with your own details. It is a standing job, so say when it should run.
Each weekday morning, read new issues in [repo] and new messages in [feedback channel]. Group recurring requests and bugs. Pick the single smallest fix that is clearly scoped, implement it in the Codex environment [name], run the test suite, and open a pull request from the bot account on a fork with a description and a recording of the change. Do not merge, deploy, change CI or secrets, or touch anything marked production. Message me the PR link and a one-line summary.
Merging, deploying and changing secrets on hand off to you. Opening a PR on take action if pre-approved. Open the dot's profile, then Customize and Custom rules, describe the action and pick the behaviour. Sensitive actions such as changing a password always stay with you.
A Codex cloud environment, the issue tracker plugin, a bot account with narrow permissions, and the dot's memory of your review feedback.
Review every PR as you would a new contributor's. Tests passing in the dot's environment do not mean the change is right, and it cannot see undocumented conventions until you tell it. Dots can make mistakes, including when following your rules; review consequential work before relying on it.
OpenAI's dots announcement and OpenAI's getting-started guide, September 29, 2026. Sundog is an independent community and is not affiliated with OpenAI; this workflow is an untested starting point.
A review queue of stale links, inconsistent examples, and missing prerequisites.