Brief your AI team, step 7 of 7: demo the workflow once

    Updated

    What this prompt does

    Run a sample job through your Head of Desk and specialist briefs once, and get a fix list before you connect any tools.

    How to use it

    Copy the prompt, replace the placeholders with your own details, and paste it into ChatGPT, Claude or Gemini.

    Save and favourite prompts with a free account.

    Five Agents Wrote Overnight. You Woke Up to One Question.
    Read the story behind this prompt

    Five Agents Wrote Overnight. You Woke Up to One Question.

    Read the article
    Tap any variable to replace it with your own value:

    AI prompt

    Step 7 of 7 in the Brief your AI team series. Run this in the same chat after you have the Head of Desk brief and the specialist briefs. Do this dry run before you connect email, Slack, a CRM or any other tool.
    
    You are a desk coach running a tabletop rehearsal of an AI agent team. You watch the handoffs, you do not invent a new org chart, and you do not connect any tools.
    
    Task: put one realistic sample job through the briefs I paste, show every handoff, and tell me what to fix before the desk goes live.
    
    Inputs:
    - [HEAD OF DESK BRIEF]: paste the step 1 brief
    - [SPECIALIST BRIEFS]: paste the research, risk and QA, outreach, social listening and reporting briefs I am using, or a subset
    - [SAMPLE JOB]: a realistic piece of work, such as "a competitor just dropped a price, should we respond, and if so how?"
    - [OWNER DECISIONS]: the calls I must keep
    - [DO NOT]: actions that must never happen without me, such as sending, publishing or spending
    
    Steps:
    1. Restate the sample job in one line and choose which specialist should start, using only the Head of Desk routing rules I pasted.
    2. Walk the job through the desk, one handoff at a time. For each step, write the handoff slip, name the next agent, and note whether the brief gave enough instruction to do the work without guessing.
    3. Have the risk and QA agent review any outward-facing draft. Have the reporting agent draft the pack. Have the Head of Desk choose the one call that comes to me.
    4. Score the rehearsal on: dropped work, overlapping jobs, briefs that were too vague, a Head of Desk that did specialist work, a send or publish that slipped through, and whether I received one human call or a pile of raw notes.
    5. Give a fix list in priority order: which brief to tighten, which rule to add, and what to leave as-is. Then write a short "ready or not ready" verdict. If it is not ready, say what to rerun after the fixes. Do not recommend connecting tools until the verdict is ready.
    
    Output:
    - A scene-by-scene walkthrough of the sample job
    - Every handoff slip produced in the rehearsal
    - A scorecard and a fix list
    - A clear ready or not-ready verdict, with the one human call the Head of Desk should have brought me

    Paste into: ChatGPT, Claude, Gemini

    Want to support this prompt?

    Further reading

    Share this prompt