Send your whole Bot team to find one improvement each
A one-line Chief of Staff prompt sends every Bot off to improve something on its own. Users say it finds good gaps, and can eat a big share of a week's usage.
Explain it simply
One short instruction tells your lead AI assistant to send every other assistant off to find one thing to improve. People say it finds useful gaps, but it can use a big share of your usage allowance, and too many assistants at once can jam their shared computer.
What was done
Debbie O'Brien shared a short prompt for a Chief of Staff Bot: gather every Bot, send each one to improve one thing on its own, be proactive, and skip anything that needs the owner's permission.
The idea is a recurring sweep that finds gaps the owner hasn't thought of.
How it went
The post reports no results, and replies are mixed. One user said it worked well but left hours of work to review.
- Usage: one reply said it used a lot of their limit; another joked it cost 40% of their usage.
- Collisions: one Chief of Staff handed out 12 jobs at once, every Bot tried to use the same computer, and it got stuck.
- One user runs it every three weeks.
- The 'nothing that needs my permission' clause carries a lot of weight.
Prompt
Get all the Bots together. Give each one a single improvement to investigate in its own area, no more than [3] at a time so they don't fight over the shared computer.
Don't send, post, buy, delete or change permissions.
For each improvement, send me a one-paragraph proposal with the expected benefit and how much usage it will take, and wait for my go-ahead before doing it.Our safer version of the original one-liner, with concurrency and usage limits.
Setup notes
- Run it from your Chief of Staff Bot, not from each Bot.
- Cap how many Bots run at once so they don't collide on the shared computer.
- Schedule it every two to three weeks, not daily.
- Check your usage meter after the first run.
Business fit
Who this helps, and how:
- Solo foundersOwners with several Bots get a periodic audit of their own operations.
- AgenciesAgencies can run it per client team to surface process gaps before reviews.
- StartupsSmall teams can use it as a fortnightly retro for their automations.
Related
Similar tools
Other tools for this job, and how they differ. Chosen by our editors, never paid for.
- Hermes Agent
Self-hosted; it saves reusable skills from work it finishes and keeps memory across sessions.
- OpenClaw
Self-hosted and built for long-running autonomous work; you maintain it and lock down what it can touch.
See it in use: Give an always-on agent a revenue target and let it run the business
Evidence grade D means: A brief, early or in-progress claim. Treat it as an idea to test, not proof. How we grade. Spotted a mistake? Suggest a correction.