Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Five standing advisors — the Operator, the Skeptic, the CFO, the Coach, the Customer — debate your decision on paper and vote. Use when asked to help me decide, pressure-test this decision, what would smart advisors say, or convene my board. Produces five distinct advisor memos, a disagreement map, a vote with a stated decision rule, and the one question to resolve before deciding.
.claude/skills/mohitagw15856-personal-board-of-directors/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 32% | 0% |
Good decisions survive five different kinds of scrutiny; most decisions get one (your own, on a good day). This skill convenes a standing board of five archetypes who each write a short memo, disagree where they honestly would, and vote — with the decision rule stated before the count.
Ask for these if not provided:
| Advisor | Lens | Their memo must contain | Their known bias (they self-disclose it) | |---|---|---|---| | The Operator | Can this actually be executed, by you, now? | The first 2 weeks, resourced honestly | Overweights logistics, underweights vision | | The Skeptic | Why this fails | The single most likely kill condition, dated | Kills some winners | | The CFO | Unit economics of the choice | Real numbers from the input, or named missing ones | Discounts the unquantifiable | | The Coach | Energy, growth, season of life | What this costs the person, not the plan | Overweights feelings | | The Customer | Would the intended beneficiary care? | The beneficiary's likely verdict, in their words | Ignores the founder's needs |
Decision rules (pick per stakes, state before voting): reversible + low stakes → simple majority · irreversible or high stakes → 4-of-5 supermajority · any advisor invokes a "stop-the-line" fact (a named, checkable disqualifier) → resolve that fact first, no vote.
Five memos, each headed by the advisor, each ending "Vote: FOR / AGAINST / ABSTAIN — flips if: named evidence]"]
| Split | Advisors | The crux | What fact resolves it | |---|---|---|---|
Decision rule: stated + why] · Count: n–n] · Board recommendation: … The one question to resolve first: the highest-leverage unknown, and how to answer it cheaply this week]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 49,124 | 30,881 | -37% | 1 | 1 | 0% | 5,907 | 4,863 | -18% | 0 | 0 | — |
case-02 | fail→pass | 41,736 | 30,334 | -27% | 1 | 1 | 0% | 4,836 | 4,572 | -5% | 0 | 0 | — |
case-03 | fail→pass | 31,153 | 29,146 | -6% | 1 | 1 | 0% | 3,385 | 4,293 | +27% | 0 | 0 | — |
case-04 | fail→pass | 25,648 | 23,383 | -9% | 1 | 1 | 0% | 2,748 | 3,524 | +28% | 0 | 0 | — |
case-05 | fail→fail | 26,010 | 32,285 | +24% | 1 | 1 | 0% | 3,872 | 4,380 | +13% | 0 | 0 | — |
case-06 | fail→fail | 35,524 | 15,305 | -57% | 1 | 1 | 0% | 4,050 | 2,248 | -44% | 0 | 0 | — |
case-07 | fail→pass | 31,621 | 34,063 | +8% | 1 | 1 | 0% | 3,381 | 4,469 | +32% | 0 | 0 | — |
case-08 | fail→pass | 50,628 | 49,348 | -3% | 1 | 1 | 0% | 3,146 | 5,063 | +61% | 0 | 0 | — |
case-09 | fail→pass | 43,933 | 31,367 | -29% | 1 | 1 | 0% | 5,453 | 4,628 | -15% | 0 | 0 | — |
case-10 | fail→pass | 39,277 | 30,453 | -22% | 1 | 1 | 0% | 2,540 | 4,772 | +88% | 0 | 0 | — |
case-11 | fail→fail | 35,433 | 29,647 | -16% | 1 | 1 | 0% | 4,258 | 4,271 | +0% | 0 | 0 | — |
case-12 | fail→pass | 34,907 | 31,714 | -9% | 1 | 1 | 0% | 3,669 | 4,187 | +14% | 0 | 0 | — |
case-13 | fail→fail | 32,539 | 34,110 | +5% | 1 | 1 | 0% | 4,490 | 4,010 | -11% | 0 | 0 | — |
case-14 | fail→pass | 18,485 | 36,344 | +97% | 1 | 1 | 0% | 1,872 | 5,418 | +189% | 0 | 0 | — |
case-15 | fail→pass | 46,256 | 30,687 | -34% | 1 | 1 | 0% | 3,887 | 4,315 | +11% | 0 | 0 | — |
case-16 | fail→pass | 26,396 | 23,803 | -10% | 1 | 1 | 0% | 2,779 | 3,592 | +29% | 0 | 0 | — |
case-17 | fail→pass | 31,471 | 42,299 | +34% | 1 | 1 | 0% | 3,014 | 6,157 | +104% | 0 | 0 | — |
case-18 | fail→pass | 62,628 | 33,711 | -46% | 1 | 1 | 0% | 8,244 | 5,053 | -39% | 0 | 0 | — |
case-19 | pass→pass | 23,323 | 42,000 | +80% | 1 | 1 | 0% | 3,133 | 6,724 | +115% | 0 | 0 | — |
case-20 | pass→fail | 16,898 | 142,759 | +745% | 1 | 1 | 0% | 2,968 | 7,430 | +150% | 0 | 0 | — |
case-21 | pass→fail | 16,386 | 10,141 | -38% | 1 | 1 | 0% | 1,337 | 2,405 | +80% | 0 | 0 | — |
case-22 | fail→pass | 21,758 | 21,461 | -1% | 1 | 1 | 0% | 3,264 | 4,613 | +41% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.