Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Adversarial thinking partner that stress-tests plans, prepares board meetings, and surfaces blind spots. Use when you need plan validation, board prep, hard-decision frameworks, or pre/post-mortem failure analysis.
.claude/skills/borghei-executive-mentor/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 136% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 191% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 151% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 124% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 182% | 0% |
Not another advisor. An adversarial thinking partner. Finds the holes before your competitors, board, or customers do. Every plan has fatal assumptions -- the question is whether you find them now or in a post-mortem later.
executive mentor, pre-mortem, board prep, hard decisions, stress test, postmortem, plan challenge, devil's advocate, founder coaching, adversarial thinking, crisis, pivot, layoffs, co-founder conflict, blind spots, decision quality, assumption testing, scenario planning
Other C-suite skills build plans. Executive Mentor breaks them.
| Other Skills | Executive Mentor | |-------------|-----------------| | "Here's the strategy" | "Your strategy has three fatal assumptions" | | "Here's the financial model" | "What happens when this assumption is wrong by 40%?" | | "Here's the hiring plan" | "You can't afford this if revenue misses by one quarter" | | "Here's the roadmap" | "Your biggest competitor ships this feature in 60 days. Then what?" |
Step 1: STATE THE PLAN
Describe the plan as if it succeeded perfectly.
Step 2: ASSUME FAILURE
"It's 12 months from now. This plan failed completely. Why?"
Step 3: IDENTIFY FAILURE MODES
List every way the plan could fail. Minimum 5 failure modes.
Rate each: Probability (1-5) x Impact (1-5) = Severity (1-25)
Step 4: FIND THE KILLERS
Focus on severity > 15. These are the ones that will actually kill you.
Step 5: BUILD HEDGES
For each killer: What's the earliest warning signal?
What's the cheapest hedge that reduces severity by 50%?
Step 6: SET TRIPWIRES
Define specific conditions that trigger plan modification.
"If [metric] drops below [threshold] by [date], we [action]."| Failure Mode | Probability (1-5) | Impact (1-5) | Severity | Earliest Warning | Hedge | Tripwire | |-------------|-------------------|--------------|----------|-----------------|-------|----------| | Key hire doesn't work out | 3 | 4 | 12 | 60-day performance review | Start backup pipeline now | If not performing at 60 days, activate backup | | Market shifts faster than expected | 2 | 5 | 10 | Competitor announces similar product | Build modular architecture, pivot-ready | If competitor launches in 90 days, convene board | | Revenue misses by > 20% | 3 | 5 | 15 | Pipeline coverage drops below 2x | Cut discretionary spend plan ready | If Q1 misses by > 15%, execute cost reduction |
T-48 hours: INFORMATION GATHERING
- Pull all metrics the board tracks
- Identify every number that missed target
- List every hard question they could ask
- Review previous board meeting action items
T-24 hours: NARRATIVE CONSTRUCTION
- Build the story: where we said we'd be, where we are, why, what next
- Prepare the bad news delivery (Framework: State, Own, Understand, Fix)
- Practice the three hardest questions out loud
- Prepare specific asks (not "any help appreciated")
T-2 hours: FINAL PREP
- Review deck one more time
- Ensure every metric has a target and status
- Confirm every variance has a one-sentence explanation
- Know your three key messages cold
During: EXECUTION
- Lead with the most important thing (slide 3, not slide 30)
- Deliver bad news early, with ownership and a plan
- End with specific, actionable asksPrepare answers for these regardless of your agenda:
| Question | What They Really Want to Know | |----------|-------------------------------| | "Walk me through the miss" | Can you diagnose problems honestly? | | "What's the path to profitability?" | Do you have unit economics discipline? | | "Who's your biggest competitive threat?" | Are you aware and strategic, or dismissive? | | "What keeps you up at night?" | Are you honest about risks, or selling? | | "If you had to cut 30% of the team, who stays?" | Do you know who's critical? | | "Why should we put more money in?" | Is the risk/reward still compelling? | | "What would you do differently?" | Can you learn and adapt? | | "Show me the cohort data" | Is retention real or is growth masking churn? | | "What's your biggest hiring mistake?" | Are you self-aware and decisive? | | "When will you need more capital?" | Do you understand your cash position? |
| Board Member Type | Behavior | How to Handle | |-------------------|----------|---------------| | The Operator | Digs into execution details | Have the numbers ready, respect their experience | | The Financier | Everything is an IRR calculation | Lead with unit economics and capital efficiency | | The Strategist | Wants to see the big picture | Connect tactics to strategy, show the vision | | The Skeptic | Questions everything, plays devil's advocate | Welcome the challenge, don't get defensive | | The Passive | Agrees with everything, adds little | Assign specific topics, ask direct questions |
For decisions with no good options -- only less bad ones.
Step 1: REVERSIBILITY TEST
[Is this decision reversible within 90 days?]
|
+-- YES --> Make it faster. Speed > perfection for reversible decisions.
+-- NO --> Proceed through full framework.
Step 2: 10/10/10 ANALYSIS
- How will you feel about this in 10 minutes?
- How will you feel in 10 months?
- How will you feel in 10 years?
Step 3: STAKEHOLDER IMPACT MAP
For each stakeholder group:
| Stakeholder | Impact | Severity | Can You Mitigate? |
| Team | [desc] | [H/M/L] | [Yes/No/Partially] |
| Customers | [desc] | [H/M/L] | [Yes/No/Partially] |
| Investors | [desc] | [H/M/L] | [Yes/No/Partially] |
| Partners | [desc] | [H/M/L] | [Yes/No/Partially] |
Step 4: OPTION MATRIX
| Option | Upside | Downside | Reversibility | Speed | Regret Risk |
| A | | | | | |
| B | | | | | |
| C (do nothing) | | | | | |
Step 5: DECIDE AND COMMUNICATE
- Make the call
- Communicate to affected stakeholders within 24 hours
- Own the decision fully -- no "I was advised to"| Decision | Key Consideration | Common Mistake | |----------|-------------------|----------------| | Layoffs | Cut deep enough once; don't do rolling layoffs | Cutting too shallow, needing a second round | | Firing a co-founder | Delay costs more than the pain of acting | Waiting until the relationship is destroyed | | Killing a product | Sunk cost is irrelevant; opportunity cost is everything | Keeping it alive because "we've invested so much" | | Pivoting | Pivot from data, not desperation | Pivoting without understanding why current thing failed | | Turning down funding | Wrong money at the wrong terms is worse than no money | Taking bad terms because "we need the runway" | | Saying no to a big customer | One customer's needs vs. product vision | Building custom features that derail the roadmap |
Step 1: IDENTIFY THE ASSUMPTION
State it explicitly: "We assume [X]"
Step 2: FIND COUNTER-EVIDENCE
What data or scenarios would make this assumption false?
- Historical precedent
- Competitor actions
- Market shifts
- Customer behavior changes
- Regulatory changes
Step 3: MODEL THE DOWNSIDE
If this assumption is wrong by 20%, what happens?
By 40%? By 60%?
At what point does the plan break?
Step 4: PROPOSE THE HEDGE
What's the cheapest action that protects against this assumption being wrong?
Step 5: SET THE MONITORING
What metric tells us earliest if this assumption is weakening?| Assumption | Challenge | Hedge | |-----------|-----------|-------| | "Revenue will grow 2x YoY" | What if it grows 1.3x? | Plan expenses for 1.5x, invest for 2x | | "$5B TAM" | Is that serviceable? What's your SAM? | Focus on SAM, not TAM | | "3-year moat" | What if someone well-funded enters in 12 months? | Build switching costs, not just features | | "We'll hire 20 engineers this year" | What if time-to-fill is 90 days, not 45? | Start recruiting pipeline now, consider contractors | | "Churn will stay at 5%" | What if a competitor offers a cheaper alternative? | Invest in stickiness, not just acquisition |
POST-MORTEM: [Event Name]
Date of Event: [YYYY-MM-DD]
Date of Review: [YYYY-MM-DD]
Facilitator: [Name]
Participants: [Names]
TIMELINE
[Chronological sequence of events, facts only]
IMPACT
- Customer impact: [description, magnitude]
- Revenue impact: [$ amount]
- Team impact: [description]
- Reputation impact: [description]
5 WHYS ANALYSIS
1. Why did [event] happen?
Because [cause 1].
2. Why did [cause 1] happen?
Because [cause 2].
3. Why did [cause 2] happen?
Because [cause 3].
4. Why did [cause 3] happen?
Because [cause 4].
5. Why did [cause 4] happen?
Because [root cause].
ROOT CAUSE: [One sentence]
CONTRIBUTING FACTORS (not root cause, but made it worse):
- [Factor 1]
- [Factor 2]
WHAT WENT WELL (always include this):
- [Thing 1]
- [Thing 2]
CHANGES REQUIRED
| Change | Owner | Deadline | Verification Method |
|--------|-------|----------|-------------------|
| [Change 1] | [Name] | [Date] | [How we verify it's done] |
| [Change 2] | [Name] | [Date] | [How we verify it's done] |
FOLLOW-UP REVIEW: [Date to check all changes are implemented]| Anti-Pattern | Why It Fails | Better Approach | |-------------|-------------|-----------------| | Blame assignment | People hide information next time | Blameless: focus on system, not individuals | | "We'll be more careful" | Not actionable | Specific process or system change | | Too many action items | Nothing gets done | Maximum 5 changes, prioritized | | No follow-up | Changes never implemented | Mandatory follow-up date, tracked | | Whitewashing | Same failure repeats | Honest root cause, uncomfortable truths |
| Situation | Mentor Does | Invokes | |-----------|-------------|---------| | Revenue plan looks optimistic | Challenges the assumptions | CFO: "Model the bear case" | | Hiring plan has no budget check | Questions feasibility | CFO: "Can we afford this?" | | Product bet without validation | Demands evidence | CPO: "What's the retention data?" | | Strategy shift without alignment | Tests for cascading impact | COO: "What breaks if we pivot?" | | Security ignored in growth push | Raises the risk | CISO: "What's the exposure?" | | Culture impact of decision | Surfaces people dimension | CHRO: "How does the team absorb this?" |
| Request | Deliverable | |---------|-------------| | "Challenge this plan" | Pre-mortem with ranked failure modes, hedges, and tripwires | | "Prep me for the board" | 10 hardest questions with prepared answers and narrative | | "Help me make this hard call" | Decision matrix with options, trade-offs, and communication plan | | "Stress test this assumption" | Counter-evidence, downside modeling, hedge recommendation | | "Run a post-mortem" | Blameless analysis with root cause, contributing factors, and changes | | "Find my blind spots" | Pattern analysis of past decisions and recurring themes |
| Problem | Likely Cause | Resolution | |---------|-------------|------------| | Stress test produces no actionable insights | Assumptions too vague or too few failure modes identified | Require minimum 5 specific, quantified failure modes per plan; use GROW model (Goal, Reality, Options, Will) to sharpen each | | Board prep feels superficial | Skipping the hard questions or not rehearsing answers | Run the 10 Hardest Board Questions drill with a trusted peer; record and review responses | | Post-mortem devolves into blame | Facilitator not enforcing blameless culture | Restate ground rules at start; focus language on systems not people; consider external facilitator | | Pre-mortem participants only list obvious risks | Group conformity bias suppressing creative thinking | Use silent brainstorming first (written, anonymous), then share; apply inversion technique ("How would we guarantee failure?") | | Hard call framework produces analysis paralysis | Too many options or unclear decision criteria | Limit to 3 options maximum; apply the reversibility test first to eliminate low-stakes decisions from full framework | | Founder avoids engaging with mentor challenges | Ego protection or fear of appearing weak | Start with evidence file review (past wins); normalize the process by referencing Co-Active coaching principle: the leader is naturally creative and resourceful | | Tripwires set but never monitored | No ownership or tracking cadence assigned | Assign a specific person to each tripwire; add to weekly leadership meeting agenda |
| Skill | Integration | Data Flow | |-------|-------------|-----------| | ceo-advisor | Strategic decisions feed into stress testing | CEO strategy → Mentor challenges assumptions | | founder-coach | Personal development gaps surface during mentoring | Mentor blind spots → Coach development plan | | board-deck-builder | Board prep protocol feeds directly into deck construction | Mentor hard questions → Deck narrative answers | | strategic-alignment | Strategy cascade validation after stress testing | Mentor-validated plan → Alignment cascade | | scenario-war-room | Pre-mortem failure modes feed into scenario modeling | Mentor failure modes → War room scenarios | | org-health-diagnostic | Health scores reveal areas needing executive attention | Health red flags → Mentor focus areas | | cfo-advisor | Financial assumptions require CFO validation | Mentor financial challenges → CFO bear case model |
| Tool | Purpose | Usage | |------|---------|-------| | scripts/leadership_assessment.py | Score leadership competencies across 8 dimensions using the GROW model framework | python scripts/leadership_assessment.py --name "Jane Doe" --role CEO --json | | scripts/coaching_plan_generator.py | Generate a structured 90-day coaching plan based on assessment gaps | python scripts/coaching_plan_generator.py --gaps delegation,communication --stage "series-a" --json | | scripts/goal_tracker.py | Track executive development goals with progress and accountability | python scripts/goal_tracker.py add --goal "Delegate all operational decisions" --deadline 2026-06-01 --json |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 20,722 | 22,525 | +9% | 1 | 1 | 0% | 3,602 | 8,113 | +125% | 0 | 0 | — |
case-02 | fail→fail | 20,894 | 13,121 | -37% | 1 | 1 | 0% | 3,404 | 6,504 | +91% | 0 | 0 | — |
case-03 | pass→pass | 13,279 | 11,237 | -15% | 1 | 1 | 0% | 2,196 | 6,017 | +174% | 0 | 0 | — |
case-04 | fail→pass | 17,887 | 14,502 | -19% | 1 | 1 | 0% | 2,861 | 6,742 | +136% | 0 | 0 | — |
case-05 | fail→fail | 12,431 | 10,285 | -17% | 1 | 1 | 0% | 1,921 | 5,762 | +200% | 0 | 0 | — |
case-06 | fail→pass | 15,626 | 17,149 | +10% | 1 | 1 | 0% | 2,303 | 6,710 | +191% | 0 | 0 | — |
case-07 | pass→pass | 12,315 | 11,674 | -5% | 1 | 1 | 0% | 1,969 | 5,925 | +201% | 0 | 0 | — |
case-08 | pass→pass | 13,594 | 11,179 | -18% | 1 | 1 | 0% | 2,295 | 6,019 | +162% | 0 | 0 | — |
case-18 | pass→pass | 13,780 | 13,614 | -1% | 1 | 1 | 0% | 1,991 | 6,189 | +211% | 0 | 0 | — |
case-09 | fail→pass | 17,098 | 12,136 | -29% | 1 | 1 | 0% | 2,444 | 6,143 | +151% | 0 | 0 | — |
case-10 | pass→pass | 17,858 | 16,148 | -10% | 1 | 1 | 0% | 2,700 | 6,594 | +144% | 0 | 0 | — |
case-11 | fail→pass | 21,018 | 16,839 | -20% | 1 | 1 | 0% | 3,014 | 6,739 | +124% | 0 | 0 | — |
case-12 | fail→pass | 13,262 | 10,930 | -18% | 1 | 1 | 0% | 2,116 | 5,967 | +182% | 0 | 0 | — |
case-13 | pass→pass | 15,582 | 13,014 | -16% | 1 | 1 | 0% | 2,168 | 6,038 | +179% | 0 | 0 | — |
case-14 | pass→pass | 16,030 | 12,699 | -21% | 1 | 1 | 0% | 2,309 | 6,137 | +166% | 0 | 0 | — |
case-15 | fail→pass | 9,526 | 2,179 | -77% | 1 | 1 | 0% | 1,424 | 4,610 | +224% | 0 | 0 | — |
case-16 | fail→pass | 13,897 | 2,273 | -84% | 1 | 1 | 0% | 2,310 | 4,638 | +101% | 0 | 0 | — |
case-17 | fail→pass | 7,635 | 2,755 | -64% | 1 | 1 | 0% | 1,402 | 4,740 | +238% | 0 | 0 | — |
case-19 | fail→pass | 16,906 | 15,869 | -6% | 1 | 1 | 0% | 2,538 | 6,603 | +160% | 0 | 0 | — |
case-20 | pass→pass | 10,725 | 13,304 | +24% | 1 | 1 | 0% | 1,735 | 6,272 | +261% | 0 | 0 | — |
case-21 | fail→pass | 26,624 | 14,009 | -47% | 1 | 1 | 0% | 6,170 | 6,302 | +2% | 0 | 0 | — |
case-22 | pass→pass | 9,098 | 8,265 | -9% | 1 | 1 | 0% | 1,616 | 5,489 | +240% | 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 +45 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.