Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Define authentic company values that guide decisions and culture — not generic corporate platitudes but actionable principles. Use when: defining company culture, creating a values document, aligning a team around shared principles.
.claude/skills/terminalskills-company-values/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 254% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 154% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 212% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 107% | 0% |
| case-17 | ✓→✓ | = Same ✓ | 41% | 0% |
Help founders and leaders define authentic company values — the foundation of culture. Based on the philosophy of The Minimalist Entrepreneur by Sahil Lavingia. Values are not generic two-word commandments; they state the non-obvious in non-obvious ways. Focus on culture before hiring: before you hire anyone, define what kind of company people want to work for.
Walk the user through these questions:
Use these as inspiration, not templates:
Help the user draft:
User prompt: "We're a 6-person remote team building an open-source database tool. We need company values before we start hiring. We believe in transparency and shipping fast."
Expected output:
Value 1: Ship It, Then Polish It We'd rather ship something imperfect today than something perfect next month. Our users are developers — they understand iteration. But "ship fast" doesn't mean "ship broken." Every release should work; it just doesn't need every feature.
Value 2: Default to Public Code is open source. Roadmaps are public. Salaries are shared internally. If you're wondering whether to share something, share it. The only exceptions: user data and security vulnerabilities.
Value 3: Builders Over Managers Everyone writes code, talks to users, or creates content. We don't have people whose only job is to coordinate other people. If a process requires a full-time manager, the process is wrong.
User prompt: "I run Simply Eloped, a small wedding elopement company. We have 12 employees across 5 states. I want values that reflect our customer-first approach and adaptability."
Expected output:
Value 1: Customer-Centric (the C in CACAO) Every decision starts with: "How does this affect the couple?" We exist to make the most important day of someone's life feel effortless. That means we answer texts at 9pm if a bride is anxious. It means we rebook a photographer when weather changes.
Value 2: Adaptable (the second A in CACAO) Elopements happen on mountaintops, in city halls, and on beaches. No two are alike. We don't follow rigid playbooks — we read the room and adjust. When COVID hit, we pivoted to virtual witnesses within a week.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 12,670 | 22,106 | +74% | 1 | 1 | 0% | 2,032 | 4,444 | +119% | 0 | 0 | — |
case-02 | fail→pass | 6,530 | 17,073 | +161% | 1 | 1 | 0% | 1,153 | 4,079 | +254% | 0 | 0 | — |
case-03 | fail→pass | 10,603 | 16,519 | +56% | 1 | 1 | 0% | 1,609 | 4,091 | +154% | 0 | 0 | — |
case-04 | fail→fail | 8,156 | 13,940 | +71% | 1 | 1 | 0% | 1,387 | 3,649 | +163% | 0 | 0 | — |
case-05 | fail→pass | 6,991 | 15,595 | +123% | 1 | 1 | 0% | 1,270 | 3,962 | +212% | 0 | 0 | — |
case-06 | fail→pass | 10,256 | 11,754 | +15% | 1 | 1 | 0% | 1,628 | 3,362 | +107% | 0 | 0 | — |
case-07 | fail→fail | 12,959 | 16,175 | +25% | 1 | 1 | 0% | 1,996 | 4,048 | +103% | 0 | 0 | — |
case-08 | fail→fail | 8,419 | 19,820 | +135% | 1 | 1 | 0% | 1,427 | 4,675 | +228% | 0 | 0 | — |
case-09 | fail→fail | 9,692 | 13,477 | +39% | 1 | 1 | 0% | 1,364 | 3,481 | +155% | 0 | 0 | — |
case-10 | fail→fail | 9,188 | 12,358 | +35% | 1 | 1 | 0% | 1,374 | 3,346 | +144% | 0 | 0 | — |
case-11 | fail→fail | 4,384 | 4,609 | +5% | 1 | 1 | 0% | 745 | 2,294 | +208% | 0 | 0 | — |
case-12 | fail→fail | 10,311 | 16,985 | +65% | 1 | 1 | 0% | 1,885 | 4,088 | +117% | 0 | 0 | — |
case-13 | fail→fail | 11,137 | 18,120 | +63% | 1 | 1 | 0% | 1,756 | 4,059 | +131% | 0 | 0 | — |
case-14 | fail→fail | 12,461 | 12,940 | +4% | 1 | 1 | 0% | 1,853 | 3,457 | +87% | 0 | 0 | — |
case-15 | fail→fail | 11,186 | 13,283 | +19% | 1 | 1 | 0% | 1,664 | 3,450 | +107% | 0 | 0 | — |
case-16 | fail→fail | 14,001 | 16,659 | +19% | 1 | 1 | 0% | 2,078 | 4,225 | +103% | 0 | 0 | — |
case-17 | pass→pass | 13,546 | 8,565 | -37% | 1 | 1 | 0% | 2,053 | 2,894 | +41% | 0 | 0 | — |
case-18 | pass→pass | 10,318 | 8,011 | -22% | 1 | 1 | 0% | 1,647 | 2,761 | +68% | 0 | 0 | — |
case-19 | pass→pass | 33,565 | 13,174 | -61% | 1 | 1 | 0% | 1,970 | 3,463 | +76% | 0 | 0 | — |
case-20 | pass→pass | 7,079 | 6,753 | -5% | 1 | 1 | 0% | 1,339 | 2,671 | +99% | 0 | 0 | — |
case-21 | pass→pass | 12,811 | 10,144 | -21% | 1 | 1 | 0% | 2,115 | 3,195 | +51% | 0 | 0 | — |
case-22 | pass→pass | 11,714 | 8,569 | -27% | 1 | 1 | 0% | 2,117 | 2,925 | +38% | 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 +18 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.