Install any skill in seconds. Free to start, no credit card required.
Get Started Free →/cs:founder-mode <question> — Auto-routes any founder question to the right C-role advisor or to /cs:boardroom for multi-role topics. The single-command entry point. Use when a founder asks any strategic question without knowing which advisor or command fits — e.g. 'runway pressure' routes to the CFO, 'gross retention dropped' routes to the CCO.
.claude/skills/alirezarezvani-founder-mode/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 106% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 176% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 57% | 0% |
Command: /cs:founder-mode <question>
The single command a founder needs to remember. Routes the question to the right C-role automatically, or triggers /cs:boardroom if multi-role.
This is the killer command — the answer to "I don't know which slash command to use." Type the question; the system figures out the room.
The router (via cs-chief-of-staff) does keyword + intent matching:
| Signal in question | Route | |---|---| | burn, runway, fundraise, dilution, model, LTV, CAC | cs-cfo-advisor | | pipeline, win rate, forecast, quota, ramp, sales motion | cs-cro-advisor | | positioning, ICP, message, brand, channel, campaign | cs-cmo-advisor | | roadmap, PMF, JTBD, North Star, RICE, kill | cs-cpo-advisor | | cadence, OKR, scorecard, DRI, operating system, rhythm | cs-coo-advisor | | hiring, comp, ladder, level, attrition, eNPS, equity | cs-chro-advisor | | security, threat, breach, compliance, audit, SOC 2 | cs-ciso-advisor | | architecture, scaling, tech debt, SLO, latency | cs-cto-advisor | | contract, IP, term sheet, regulator, license | cs-general-counsel-advisor | | retention, GRR, NRR, churn, customer success, CSM, time-to-value, renewals | cs-cco-advisor | | training data, data rights, consent, data asset, warehouse, lakehouse, data mesh | cs-cdo-advisor | | model selection, eval, hallucination, AI risk, EU AI Act, fine-tune, build vs buy AI | cs-caio-advisor | | DORA, cycle time, deploy frequency, eng hiring funnel, team topology, delivery throughput | cs-vpe-advisor | | strategy, vision, board, M&A, raise, exit | cs-ceo-advisor | | 2+ signals from different roles | /cs:boardroom | | ambiguous | /cs:office-hours first, then route |
/cs:brief and trigger /cs:boardroom/cs:office-hours to force the founder to sharpendecision-loggerThe router emits one of three responses:
**Routing:** cs-cfo-advisor
**Why:** Question hits burn rate and unit economics.
**Next:** Invoking cs-cfo-advisor with company-context loaded.
[Advisor's response follows]**Routing:** /cs:boardroom
**Why:** Question touches CFO + CMO + CPO (pricing change has finance, positioning, and product implications).
**Next:** Building brief via /cs:brief, then running boardroom.
Brief saved: ~/.claude/briefs/2026-05-12-pricing-v3.md
Run: /cs:boardroom ~/.claude/briefs/2026-05-12-pricing-v3.md**Routing:** /cs:office-hours
**Why:** Question is too broad ("should we grow faster?"). Need framing before any advisor can help.
**Next:** Six-question intake.
[Office hours questions follow]gstack requires the founder to know all 23 slash commands and pick the right one. That's a cognitive tax. /cs:founder-mode collapses that to one — the system picks. This is also where persistent memory pays off: with company-context.md + decision-logger, the router knows what's already been decided and won't re-litigate.
/cs:founder-mode "should we raise a Series B now or wait 6 months?"
→ boardroom (CFO + CEO + CRO touched)
/cs:founder-mode "the win rate dropped 20% this month"
→ cs-cro-advisor
/cs:founder-mode "gross retention dropped 5 points this quarter"
→ cs-cco-advisor
/cs:founder-mode "let's hire a VP Marketing"
→ boardroom (CHRO + CMO + CFO touched)
/cs:founder-mode "should we be growing faster?"
→ /cs:office-hours (too ambiguous)cs-chief-of-staff — does the routingchief-of-staff — routing logiccontext-engine — loads contextVersion: 1.0.0
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,767 | 17,913 | -14% | 1 | 1 | 0% | 2,305 | 3,324 | +44% | 0 | 0 | — |
case-02 | fail→pass | 14,109 | 9,808 | -30% | 1 | 1 | 0% | 1,415 | 2,214 | +56% | 0 | 0 | — |
case-03 | fail→pass | 17,355 | 18,205 | +5% | 1 | 1 | 0% | 1,693 | 3,488 | +106% | 0 | 0 | — |
case-04 | fail→pass | 12,037 | 13,307 | +11% | 1 | 1 | 0% | 936 | 2,588 | +176% | 0 | 0 | — |
case-05 | fail→pass | 19,056 | 17,838 | -6% | 1 | 1 | 0% | 1,963 | 3,086 | +57% | 0 | 0 | — |
case-06 | fail→pass | 22,027 | 25,417 | +15% | 1 | 1 | 0% | 2,362 | 3,336 | +41% | 0 | 0 | — |
case-07 | fail→pass | 28,599 | 18,998 | -34% | 1 | 1 | 0% | 2,648 | 3,253 | +23% | 0 | 0 | — |
case-08 | fail→pass | 25,214 | 34,438 | +37% | 1 | 1 | 0% | 2,932 | 3,336 | +14% | 0 | 0 | — |
case-09 | fail→pass | 18,812 | 18,052 | -4% | 1 | 1 | 0% | 2,465 | 3,240 | +31% | 0 | 0 | — |
case-10 | fail→pass | 20,956 | 15,420 | -26% | 1 | 1 | 0% | 2,483 | 3,756 | +51% | 0 | 0 | — |
case-11 | fail→pass | 24,411 | 18,227 | -25% | 1 | 1 | 0% | 2,513 | 3,434 | +37% | 0 | 0 | — |
case-12 | fail→pass | 16,283 | 15,366 | -6% | 1 | 1 | 0% | 2,585 | 2,912 | +13% | 0 | 0 | — |
case-13 | fail→pass | 18,525 | 18,997 | +3% | 1 | 1 | 0% | 2,195 | 3,427 | +56% | 0 | 0 | — |
case-14 | fail→pass | 21,061 | 23,454 | +11% | 1 | 1 | 0% | 3,096 | 3,873 | +25% | 0 | 0 | — |
case-15 | fail→pass | 19,629 | 11,695 | -40% | 1 | 1 | 0% | 2,062 | 3,222 | +56% | 0 | 0 | — |
case-16 | fail→fail | 17,663 | 12,595 | -29% | 1 | 1 | 0% | 1,835 | 3,244 | +77% | 0 | 0 | — |
case-17 | fail→pass | 18,496 | 13,679 | -26% | 1 | 1 | 0% | 2,645 | 3,398 | +28% | 0 | 0 | — |
case-18 | fail→pass | 58,013 | 14,435 | -75% | 1 | 1 | 0% | 2,104 | 2,701 | +28% | 0 | 0 | — |
case-19 | fail→fail | 15,151 | 11,894 | -21% | 1 | 1 | 0% | 2,142 | 2,340 | +9% | 0 | 0 | — |
case-20 | pass→fail | 16,417 | 20,256 | +23% | 1 | 1 | 0% | 1,928 | 3,660 | +90% | 0 | 0 | — |
case-21 | pass→fail | 19,173 | 15,489 | -19% | 1 | 1 | 0% | 2,448 | 3,622 | +48% | 0 | 0 | — |
case-22 | pass→fail | 11,397 | 15,670 | +37% | 1 | 1 | 0% | 1,331 | 3,087 | +132% | 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 +64 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/10/2026 | +76% |
Other measured skills in the registry, with their headline benchmark lift.