Coordinate long coding jobs with a chief-of-staff agent
- Level Intermediate
- Setup1–3 hrs (estimated)
- Evidence Anecdotal
- CostNot stated in source
- No-codeNo
What they did
One Claude Code session writes briefs, checks claims and reads diffs while separate sessions do the work, with an external task board holding state so nothing is lost when context compacts. Every agent report is re-run before it's believed.
How it went
Detailed blog with loop, tooling and failure modes
Worth knowing
- Detailed blog with loop, tooling and failure modes.
Prompt
Cleaned version rewritten from the source — not a direct quote.
You are the coordinator. Don't write code yourself. Break [project] into tasks on [board], write a short brief for each worker session, and when a worker says it's done, re-run its tests and read the diff before marking it complete. Keep all state on the board, not in this chat.
Recommended
Try it in Grok Bot
- Copy the prompt (button below).
- Open Grok Bot chat and paste it. Replace any [bracketed] placeholders.
- Connect the listed tools (as needed) as connectors when Grok Bot asks.
- If the task repeats, ask: "Make this a daily routine."
Alternatives: the same prompt works in Claude Code, Claude, ChatGPT Agent, OpenClaw.
Set it up
Stack: Claude Code
- Read the original source for full context.
- Connect the tools your agent needs.
- Paste the prompt and replace [bracketed] placeholders.
- Run once manually and check the output before scheduling it.
Generic suggested steps, not from the source.
Evidence
Detailed blog with loop, tooling and failure modes
Source: Web · Verified by us: not yet · Report broken or outdated