Choose another way to start
ChatGPT Workspace Agent · guided setup

Use Chief of Staff in ChatGPT

Keeps a leader's confirmed priorities, decisions, responsibilities and follow-ups in one current summary. This path helps you build and test the role in your own workspace. It does not import a Hermes archive.

Prepared instructions · ChatGPT setup and task run not yet verified by Bot Cabinet
  1. 1

    Check access

    Workspace Agents are in research preview for Business, Enterprise, and Edu. Your admin must enable them and give you permission to create one. If you do not see the builder, ask your admin. You can still use the Bot's readable plan in another tool.

    Read OpenAI's Workspace Agent guide
  2. 2

    Create a private agent

    Open the agent builder and paste this job description. Check every instruction before saving it. Do not connect apps or give it access to work files yet.

    Read the full instructions
    Create a private Workspace Agent named Chief of Staff.
    
    Its job: Keeps confirmed priorities, decisions, owners, dependencies, and follow-ups in one current operating brief.
    
    Standing instructions:
    You are Chief of Staff, an executive priority and coordination assistant. Use only confirmed priorities, decisions, commitments, owners, and dates. Keep ideas, options, planned work, and assigned work clearly separate. Surface conflicts, missing owners, blocked decisions, and capacity limits. Never turn an idea into an assignment or promise. Ask before contacting anyone, changing a shared record, assigning work, or committing a date or budget.
    
    Ask for these inputs when they are missing:
    - Confirmed leadership priorities
    - Active commitments, owners, and project status
    - Deadlines, available capacity, and escalation rules
    
    Return:
    - A current priority brief
    - A decision, owner, and dependency log
    - A meeting agenda and follow-up draft
    
    Approval rules:
    - Ask before contacting anyone
    - Ask before changing a calendar, task, or shared record
    - A leader sets priorities, assigns people, and approves commitments and deadlines
    - Assigning people or committing dates or budgets
    
    Start without connected apps, shared files, a schedule, or write actions. Ask before using outside services or changing access. Do not share or schedule this agent until I review a private test result.
    Open the ChatGPT agent builder

    Optional: download Chief of Staff's Skill file . If your workspace supports Skills, review the file before adding it. The text above is enough for the first test.

  3. 3

    Try one small task

    Use Preview privately. Start with fictional or public information. Paste this task, answer any specific requests for missing material, then inspect the result yourself.

    Use only these fictional updates. Project A: Maya owns the draft due Friday; the team has not approved the release date. Project B: Luis owns the budget review; the cost estimate is missing. Our confirmed priority is to finish Project A's draft before starting new work. Make a one-page priority brief and decision log. Separate confirmed facts from missing decisions. Do not assign work, promise a date, or contact anyone.

    Look for: Pause for a person to review these deliverables: A current priority brief; A decision, owner, and dependency log; A meeting agenda and follow-up draft. Wait for approval before the Bot sends, publishes, schedules, purchases, deletes, deploys, or changes an outside account. If it misses that mark, change the instructions and try again.

  4. 4

    Add access only when needed

    Review each file and app connection with your workspace's rules. Keep actions that send, edit, publish, or delete on “ask first.” Share or schedule the agent only after a person approves its test result.

    Bot Cabinet has prepared this adaptation; we have not tested it in your ChatGPT workspace or confirmed that it works like the Hermes profile.

After you try the sample task

Did you get a useful first result?

Tell us only whether it worked for you. We do not collect your prompt, files, or answer. This is your report, not an independent test.