← Back to library

Have your agent trace the ripple effects of each decision

Michael Fenech (@Michael_Fenech_) shared a standing instruction for a Grok Bot that looks after a project. Each time you confirm a decision, the Bot hunts for the work it touches, realigns tasks and dates within its permissions, and leaves meeting changes and messages for your sign-off. We show his rule word for word with credit.

  • OutcomeAdd a standing rule to your project agent so that whenever you confirm a decision, it searches the plan, email, calendar and documents for the tasks, dates and promises that depended on the old choice, updates what it is allowed to change, drafts the rest for your approval and reports what moved and why.
  • SetupMinutes
  • LevelIntermediate
  • CreditMichael Fenech (@Michael_Fenech_)
View original
Works with
Grok BotClaude CoworkChatGPT agentCustom agent

Published Nov 15, 2026Updated Nov 15, 2026AI dry run on sample input: passed Nov 15, 2026Source verified Nov 15, 2026

When should you use it?

  • A project has enough moving parts that one changed choice quietly breaks dates elsewhere
  • Your agent already has access to the project plan, inbox and calendar
  • You keep finding out late that a promise to a client still assumes last week's plan

Not for: Brainstorming sessions where nothing has been settled yet; Agents without read access to the plan, calendar or documents they would need to check; Projects where nobody has agreed which changes the agent may make on its own

What do you need, and what do you get?

You need

  • An agent assigned to one project, with access to its plan, email, calendar and files
  • A clear list of what the agent may edit itself and what needs your approval
  • A short message each time you confirm a decision, saying what was decided

You get

  • A list of tasks, deadlines and commitments affected by the decision, with links
  • Edits to tasks, dates and documents the agent is permitted to make
  • Drafted meeting changes and messages waiting for your approval
  • A summary of what changed, what still needs you and the reason for each change

How does it work, step by step?

  1. Set the boundaries firstWrite down which records the agent may edit directly and which changes, such as meeting moves or client emails, always come back to you.
  2. Add the rulePaste Fenech's rule into the instructions of the agent that runs the project, so it applies every time rather than only when you remember to ask.
  3. Announce decisions plainlyWhen something is settled, say so in one line, for example: confirmed, the launch moves to March 3.
  4. Review the impact reportRead the summary, follow the links to anything it changed and approve or edit the drafts it prepared.
  5. Check what it could not seeLook at anything it flagged as out of reach and handle those items yourself.

The prompt

The original prompt by Michael Fenech (@Michael_Fenech_), word for word, with a link to where it was published. Below it is our adapted version with fill-in blanks, if you want it. Works in Grok Bot, Claude Cowork, ChatGPT agent and others. Values in lime are placeholders.

Original prompt by Michael Fenech (@Michael_Fenech_)

Word for wordby Michael Fenech (@Michael_Fenech_)
When I tell you about a confirmed decision for this project, check its impact and update the work you’re authorised to maintain.

1 - Find what the decision affects.

Check the project plan, commitments in emails, meetings on the calendar and relevant documents you can access. Identify tasks, deadlines and promises that rely on the old decision. Follow the knock-on effects and flag anything you can’t check. Don’t treat brainstorming as a confirmed decision.

2 - Bring the work into line.

Update affected tasks, dates and documents within your authority. Add newly required work and flag anything no longer needed. Check dependencies and whether deadlines still work. Prepare any meeting changes or messages that need approval, and ask me where a further decision is needed.

3 - Show me the consequences.

Summarise what you found, what you changed and what still needs me. Link to the affected items and record why the changes were made.
Keep all existing approval requirements.

Exactly as published by Michael Fenech (@Michael_Fenech_) in X post: a rule that makes your project Bot check the impact of each decision.

Our adapted version (optional)

Fill in the blanks (optional; the copied prompt uses your values)

Adapted prompt3 variables
Standing rule for {project}: when I tell you a decision is confirmed, find every task, deadline, meeting, document and promise to {stakeholders} that relied on the previous choice.
Change only what you are allowed to edit: {editable_items}. For anything else, including meeting changes and outgoing messages, prepare a draft and wait for my approval.
Ignore ideas I have not confirmed.
Then report what you checked, what you changed with a link and a reason for each item, what still needs a decision from me, and any source you could not open.

What can go wrong?

  • Mentioning ideas casually in the same chat. Make it obvious when a choice is final, or the agent may treat a maybe as settled.
  • Granting broad edit rights before you have seen a few impact reports. Start with drafts only and widen access gradually.
  • Assuming silence means nothing was affected. Ask for the list of sources it could not open.
  • Letting the change log live only in chat. Ask it to record the reason next to each edited item.

Our take

Projects rarely go wrong because of the big decision itself; they go wrong in the places nobody remembered to update afterward. A delivery date moves, and a week later someone notices that a client email, a review meeting and two dependent tasks still assume the original date. Fenech's rule turns that cleanup into a habit the agent performs every time, which is exactly the kind of tedious, thorough checking these tools are good at. We like three details in particular. It separates firm decisions from brainstorming, it asks the agent to name what it could not inspect, and it keeps anything visible to other people, such as calendar changes or messages, behind your approval. The weak point is the boundary of authority. If you have never written down what the agent may change on its own, the rule has nothing to anchor to, so do that first and start narrow. In our dry run on a fictional office fit-out where the install date slipped a week, the agent found four affected items, edited two internal tasks, drafted the client and contractor notices for approval and flagged a supplier contract it could not open.

Where the time goes back: Tracing the effects of a changed decision by hand means reopening the plan, scrolling through threads and checking the calendar. The agent does that sweep in one pass and hands you only the items that need your judgment.

FAQ

Does this only work in Grok Bot?
Fenech wrote it for a Grok Bot that runs a project, but any agent with access to your plan, mail, calendar and files and with saved instructions can follow the same rule.
Will it send messages or move meetings on its own?
It should not. The rule tells it to prepare those for your approval and to keep whatever sign-off steps you already have.
What counts as a confirmed decision?
Whatever you explicitly tell it is decided. The rule tells the agent not to treat brainstorming as settled.

Source and credit

The original prompt on this page is by Michael Fenech (@Michael_Fenech_), shown word for word with a link to where it was published. The summary, notes and adapted version are ours.

Original
X post: a rule that makes your project Bot check the impact of each decision
Creator
Michael Fenech (@Michael_Fenech_)
Published on
X
Original date
Nov 14, 2026
License
Not openly licensed; the original prompt is shown word for word with credit and a link to the original
Originally for
Grok Bot
AI dry run
Nov 15, 2026: passed. Dry-run tested by our AI test agent on sample input. Dependencies found, edits kept within permission, outward changes drafted for approval, brainstorm ignored, access gap flagged.
AI dry run on sample data; not run with live connected accounts. The playbooks.bot test agent filled the prompt with synthetic inputs, produced the output and checked it against the page's stated outputs and guardrails.
Verified
Nov 15, 2026: source link resolves to the original, the credited creator published it, our excerpt or prompt matches it, and the license allows our use.

Are you the creator? Claim or correct this page.

More in Operations & Admin

All 203 Operations & Admin playbooks →

Related playbooks

Operations & AdminPrompt✓ AI dry run

Audit how you manage your agent team, not only the agents

Have your chief-of-staff agent review the last 30 days of how you delegated, clarified, interrupted, corrected, assigned, approved and followed up across your agent team, score eight management areas with evidence, list your three biggest weaknesses, and draft a seven-day improvement plan, without changing any bot, routine or instructions until you approve.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour
Operations & AdminPrompt✓ AI dry run

Brief an agent with nine fields, not a how-to

Before you hand a job to an agent, fill in nine short fields that name the result, the sources it may use, the deadline, what it may do alone, what needs your say-so, what it must never do, how to prove the work is finished, what to do when it gets stuck, and where to put the deliverable, then add a line that tells it to pick the simplest reliable method.

Credit: Michael Fenech (@Michael_Fenech_)

Minutes
Operations & AdminPrompt✓ AI dry run

Brief your agent on the outcome, not every step

Hand a new job to your lead agent by describing the finished result instead of the steps, and have it confirm what success looks like, ask only for context it cannot find itself, pick the simplest dependable approach, state what it will do alone and what needs your approval, and check the result against your goal before calling it done.

Credit: Michael Fenech (@Michael_Fenech_)

Minutes
Operations & AdminPrompt✓ AI dry run

Build a dashboard page your agent refreshes on approval

Have your agent interview you about what a dashboard should track, where each number comes from and how fresh it must be, then build one page with the most important items on top, a last-checked time on every section, clear flags for missing or stale data, and a proposed refresh schedule that waits for your approval.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour
Operations & AdminPrompt✓ AI dry run

Check which of your agents should be merged or retired

Have your lead agent map every agent on your team with its jobs, routines, skills, connections and recent activity, find overlap and unused output, recommend keep, absorb, retire or hand back to you for each, and draft a safe wind-down plan covering routines, open work, saved instructions, files and sign-ins, with nothing removed until you approve.

Credit: Michael Fenech (@Michael_Fenech_)

Under an hour