feat: resilient multi-channel messaging - #641
Conversation
Engineer prompt now instructs workers to call both `genie done` (state) and `genie send` (mailbox notification) on completion. Reviewer, QA, and fix prompts add `genie send` for durable result reporting to team-lead. Flat copies synced.
- Update initialPrompt in dispatch.ts to include `genie done` and `genie send` commands with actual slug/group values (Group 2) - Update team-lead prompt with state-polling-first monitoring: `genie status` as primary, `genie inbox` as secondary, "state file is source of truth" (Group 3) - Sync team-lead flat copy
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly improves the robustness and clarity of inter-agent communication and status tracking within the Highlights
Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces a more resilient multi-channel messaging system for agent completion reporting. Worker agents now use both genie done to update a state file (the source of truth) and genie send to send a durable notification to the team-lead. The prompts for all agents (engineer, reviewer, qa, fix, team-lead) have been updated to reflect these new procedures, and the dispatch logic in dispatch.ts has been updated accordingly. The changes are well-aligned with the goal of improving messaging resilience. I've suggested a minor improvement for consistency in the engineer agent's prompt.
| 1. Run validation commands from the wish | ||
| 2. Commit and push your work | ||
| 3. Call: `genie done <slug>#<group>` — marks the group complete in state (source of truth) | ||
| 4. Call: `genie send 'Group <N> complete. <summary>' --to team-lead` — sends durable notification |
There was a problem hiding this comment.
For consistency and to reduce ambiguity for the agent, it would be better to use <group> here instead of <N>. The genie done command on the preceding line uses <group>, and the corresponding initialPrompt in dispatch.ts is constructed with the actual group number. Using <group> makes it explicit which value should be substituted.
| 4. Call: `genie send 'Group <N> complete. <summary>' --to team-lead` — sends durable notification | |
| 4. Call: `genie send 'Group <group> complete. <summary>' --to team-lead` — sends durable notification |
| 1. Run validation commands from the wish | ||
| 2. Commit and push your work | ||
| 3. Call: `genie done <slug>#<group>` — marks the group complete in state (source of truth) | ||
| 4. Call: `genie send 'Group <N> complete. <summary>' --to team-lead` — sends durable notification |
There was a problem hiding this comment.
For consistency and to reduce ambiguity for the agent, it would be better to use <group> here instead of <N>. The genie done command on the preceding line uses <group>, and the corresponding initialPrompt in dispatch.ts is constructed with the actual group number. Using <group> makes it explicit which value should be substituted.
| 4. Call: `genie send 'Group <N> complete. <summary>' --to team-lead` — sends durable notification | |
| 4. Call: `genie send 'Group <group> complete. <summary>' --to team-lead` — sends durable notification |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 8aeaab5fe2
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| After completing all deliverables and validation, call `genie done <slug>#<group>` to report completion. The slug and group are in your initial prompt. This signals the orchestrator that your work is finished so it can advance to the next wave. | ||
| After completing all deliverables and validation: | ||
| 1. Run validation commands from the wish | ||
| 2. Commit and push your work |
There was a problem hiding this comment.
Making Commit and push your work a required engineer step can stall normal orchestration: workers now have to push before calling genie done, so any push failure (no remote access, auth issues, branch protection) prevents state completion and causes genie work to hang until timeout, and it also conflicts with the team-lead flow that already performs the canonical git add/commit/push in Phase 3 (plugins/genie/agents/team-lead/AGENTS.md lines 36-40).
Useful? React with 👍 / 👎.
1 participant
Summary
genie doneandgenie sendgenie status) as source of truth, with inbox as secondaryinitialPromptin dispatch.ts to includegenie doneandgenie sendcommands with actual slug/group valuesWish
resilient-messaging
Test plan
bun run checkpasses (751 tests, typecheck, lint, dead-code)bun run buildpasses (0.36 MB bundle)genie doneandgenie sendin initialPromptgenie statusbeforegenie inbox