Advisory Board
Compose a personal board of advisors, pressure-test any real decision with structured dissent, and leave with a grounded recommendation and a memo ready for your team.
What it is
Section titled “What it is”The advisory board workbench lets you compose a personal board of advisors – respected figures, domain thinkers, or personal mentors you choose – and convene them on real decisions.
A coordinator builds the decision story with you before anyone speaks. The board then deliberates from distinct lenses. A standing “10th Man” seat is always at the table and is obligated to argue against whatever consensus the board reaches, surface the black swan, and hand you tripwires to watch. A neutral clerk (the yeoman) records the deliberation, applies a fitting decision structure, and compiles a memo you review and approve before it goes anywhere.
The coaching is the thinking. The memo is the deliverable.
When to use it
Section titled “When to use it”Reach for the advisory board when a decision carries real stakes: a contract you are about to sign, a hire, a pricing call, a strategic pivot, a significant spend. The value is in the structured prep and the dissent – not speed.
If the decision is cheap and easy to undo, skip the board and decide.
What you leave with:
- A clearer framing of the real decision (your first question is rarely the right one)
- Deliberation from advisors reasoning from distinct, genuine lenses
- A mandatory opposing case from the 10th Man, with tripwires to monitor
- A recommendation weighted by your values, not a vote count
- A ready-to-approve decision memo for your team
How to use it
Section titled “How to use it”The workbench has two commands. Use them in order: setup once to compose your board, then convene each time you bring a decision.
Setting up your board
Section titled “Setting up your board”Type setup to your CLI agent. The coordinator opens with questions that surface who you trust and why – this is intentional. Choosing your board distills your values. The coordinator guides but never derives it for you.
What setup covers:
- Eliciting advisors through open questions (“Whose judgment do you trust when you are about to make a mistake?” / “Who would you actually want to tell you you are wrong?”). Aim for 3-5 advisors you are genuinely excited about.
- Building each persona on selection. For a documented figure, the coordinator draws on their real body of work. For a personal mentor, it interviews you. Every persona is built from scratch for your board – there is no pre-loaded library.
- A mandatory confirmation gate. Before any persona is seated, the coordinator shows you the draft and asks whether it matches the advisor you respect. You edit until it does. An unconfirmed persona is never seated.
- The 10th Man, seated automatically. It is always on your board. You may rename or theme it. You cannot remove it.
- An optional spiritual seat. The coordinator asks once, without assumption, whether you factor faith or spiritual considerations into decisions. If yes, it builds a seat grounded in your stated tradition, marked as standing (it weighs in on every convene as a values lens, not only on spiritual questions). If no, the subject is dropped.
- A mirror. The coordinator reflects the board’s shape back to you – gaps, balance, composition – and asks whether it is what you want. You make every call.
Your board is saved. Run setup again at any time to add a seat or refine a persona.
Convening the board
Section titled “Convening the board”Type convene to your CLI agent and bring a decision. The coordinator will not convene the board until the decision story is solid.
Step 1: Build the story (the prep). The coordinator interrogates you into a clear decision brief. It captures:
- The decision stated precisely, with the real options (not just yes/no)
- The goal the decision actually serves
- What is known versus what is assumed, separated explicitly
- Stakes and reversibility – what is lost if it is wrong, and whether it can be undone
- Your own lean and why – so the board challenges it rather than guessing at it
- The unknowns you are glossing over
The coordinator reflects the brief back and waits for your confirmation before continuing. This step is roughly 80% of the eventual memo.
Step 2: Deliberation. Your seated advisors weigh in on the brief in their own voice and from their own lens. The coordinator pursues genuine tension, not agreement theater. Each advisor ends with their lean.
Step 3: The 10th Man (runs last). After the board’s takes are in, the 10th Man reads the emerging consensus and produces, in earnest:
- The strongest opposing case the board is dismissing
- The black swan scenario, treated as if it will happen
- What would have to be true for that scenario
- Tripwires: leading indicators framed as kill-clauses (“if this moves, the consensus was wrong”)
The stronger the board’s consensus, the harder it works. On a genuinely split decision it is quiet. On a unanimous call it goes to maximum.
Step 4: Synthesis. The coordinator reconciles the board into a recommendation – not a vote tally. Where advisors conflict, the tiebreak is your foundation values. The right output is often not yes/no but a cheaper next move that resolves the board’s core disagreement.
Step 5: The memo. The yeoman (an opinion-neutral clerk that never sat in deliberation) records what was said, selects the fitting decision structure from its toolkit, and compiles the memo:
Decision: <precisely stated, with the real options>Context: <the known facts; key assumptions flagged>
Recommendation: <the synthesis: the call + the immediate next action>
Decision method: <the structure applied> -> <its result>
How the board weighed in:- <seat>: <one-line lean and why>- ...
10th Man (tail risk): <the black swan taken seriously, one short paragraph> Tripwires: <leading indicators / kill-clauses> Any two -> pause and re-decide.
Next action: <the single concrete next step>The yeoman drafts. You approve before the memo goes to your team.
For high-stakes decisions, ask your assistant for a deep convene: each seat runs as an independent agent rather than simulated in one pass. Independence reduces groupthink at the cost of time and tokens. Reserve it for decisions that warrant it.
Commands at a glance
Section titled “Commands at a glance”| Command | What it does |
|---|---|
setup |
Compose or update your board |
convene |
Take a decision to the board |
status |
Show setup state and the list of seated advisors |
Common pitfalls
Section titled “Common pitfalls”A homogenized board. If your advisors share the same worldview, deliberation collapses into consensus theater. The 10th Man is your structural defense, but it works better against a board that already disagrees. When the coordinator mirrors the board’s composition back to you at setup, take the gap observations seriously: heavy on strategy, light on operators. Three contrarians and no one who has run payroll – those gaps matter.
Ignoring the 10th Man findings. The tripwires in the tail-risk block are not boilerplate. They are early-warning indicators the 10th Man identified as kill-clauses: two of them moving means the consensus was wrong and you should pause. Build them into your monitoring. A memo without tripwires acted on is a memo you approved but did not use.
No foundation values, weak synthesis. Where advisors conflict, the coordinator resolves it by weighting toward your foundation values and philosophy. Without a foundation setup, that weighting is generic and the synthesis will feel flat or arbitrary. If recommendations feel like they could belong to anyone, the missing input is your documented values.
Arriving with a fuzzy question.
“Should I grow?” is not a decision the board can pressure-test. The coordinator will push you to reframe, but you move faster if you rough out the real options before you type convene. Instead of “grow or not,” bring “expand to a second service line vs. deepen the current one vs. hold six months and revisit.”
Shipping the memo raw. The yeoman drafts. You approve. The board sharpens your position – it does not supply one for you. Read the memo and make it yours before it goes to anyone.
Troubleshooting
Section titled “Troubleshooting”| Symptom | Likely cause | Fix |
|---|---|---|
| Board reports “status: UNCONFIGURED” | Setup has not been completed | Run setup. Seat at least one advisor and confirm the 10th Man is present |
| Coordinator will not start the deliberation | The decision brief is still too fuzzy | Continue the story-building exchange. The coordinator holds until the brief is confirmed |
| All advisors reach the same conclusion immediately | Board composition is too homogenized | Re-examine the board at setup. Add a seat with a genuinely contrarian or different-domain lens |
| The 10th Man output seems brief | The board was already split before it ran | Normal behavior – the 10th Man scales with the degree of consensus it sees |
| Synthesis feels generic or arbitrary | Foundation values are not set | Set up your foundation workbench so the coordinator has your actual values and philosophy to weight by |
| Memo is missing the tail-risk section | The convene was cut short before the 10th Man ran | Re-run convene. Do not let the yeoman compile until Step 3 is complete |
| A persona feels off or inaccurate | The confirmation gate was rushed | Run setup again, re-examine that persona, and edit until it matches the advisor you actually respect |
| Synthesis recommends something you cannot act on | Options were under-specified in the brief | Re-run convene with sharper options. Include concrete alternatives and the constraints that bound them |