Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →- Before starting any feature — validate the "why"
.claude/skills/loulanyue-product-lens-think-before-you-build/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -30% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 37% | 0% |
Like YC office hours but automated. Asks the hard questions:
1. Who is this for? (specific person, not "developers")
2. What's the pain? (quantify: how often, how bad, what do they do today?)
3. Why now? (what changed that makes this possible/necessary?)
4. What's the 10-star version? (if money/time were unlimited)
5. What's the MVP? (smallest thing that proves the thesis)
6. What's the anti-goal? (what are you explicitly NOT building?)
7. How do you know it's working? (metric, not vibes)Output: a PRODUCT-BRIEF.md with answers, risks, and a go/no-go recommendation.
Reviews your current project through a founder lens:
1. Read README, CLAUDE.md, package.json, recent commits
2. Infer: what is this trying to be?
3. Score: product-market fit signals (0-10)
- Usage growth trajectory
- Retention indicators (repeat contributors, return users)
- Revenue signals (pricing page, billing code, Stripe integration)
- Competitive moat (what's hard to copy?)
4. Identify: the one thing that would 10x this
5. Flag: things you're building that don't matterMaps the actual user experience:
1. Clone/install the product as a new user
2. Document every friction point (confusing steps, errors, missing docs)
3. Time each step
4. Compare to competitor onboarding
5. Score: time-to-value (how long until the user gets their first win?)
6. Recommend: top 3 fixes for onboardingWhen you have 10 ideas and need to pick 2:
1. List all candidate features
2. Score each on: impact (1-5) × confidence (1-5) ÷ effort (1-5)
3. Rank by ICE score
4. Apply constraints: runway, team size, dependencies
5. Output: prioritized roadmap with rationaleAll modes output actionable docs, not essays. Every recommendation has a specific next step.
Pair with:
/browser-qa to verify the user journey audit findings/design-system audit for visual polish assessment/canary-watch for post-launch monitoring| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | fail→fail | 10,545 | 2,533 | -76% | 1 | 1 | 0% | 1,434 | 1,021 | -29% | 0 | 0 | — |
case-01 | fail→pass | 9,163 | 6,774 | -26% | 1 | 1 | 0% | 1,484 | 1,721 | +16% | 0 | 0 | — |
case-02 | fail→fail | 8,949 | 5,308 | -41% | 1 | 1 | 0% | 1,371 | 949 | -31% | 0 | 0 | — |
case-03 | fail→pass | 9,016 | 7,479 | -17% | 1 | 1 | 0% | 1,356 | 1,714 | +26% | 0 | 0 | — |
case-04 | pass→pass | 9,755 | 8,350 | -14% | 1 | 1 | 0% | 1,529 | 1,792 | +17% | 0 | 0 | — |
case-05 | fail→pass | 11,168 | 7,468 | -33% | 1 | 1 | 0% | 1,726 | 1,811 | +5% | 0 | 0 | — |
case-06 | pass→pass | 12,527 | 10,441 | -17% | 1 | 1 | 0% | 1,769 | 2,164 | +22% | 0 | 0 | — |
case-07 | pass→pass | 13,519 | 12,862 | -5% | 1 | 1 | 0% | 2,210 | 2,647 | +20% | 0 | 0 | — |
case-08 | fail→pass | 18,268 | 7,734 | -58% | 1 | 1 | 0% | 2,787 | 1,949 | -30% | 0 | 0 | — |
case-10 | pass→pass | 17,992 | 11,999 | -33% | 1 | 1 | 0% | 2,810 | 2,536 | -10% | 0 | 0 | — |
case-11 | fail→pass | 13,716 | 14,691 | +7% | 1 | 1 | 0% | 2,122 | 2,897 | +37% | 0 | 0 | — |
case-12 | fail→fail | 9,276 | 5,709 | -38% | 1 | 1 | 0% | 1,438 | 1,537 | +7% | 0 | 0 | — |
case-13 | pass→pass | 11,460 | 2,521 | -78% | 1 | 1 | 0% | 1,761 | 1,042 | -41% | 0 | 0 | — |
case-14 | fail→fail | 9,590 | 2,142 | -78% | 1 | 1 | 0% | 1,457 | 993 | -32% | 0 | 0 | — |
case-15 | pass→pass | 18,085 | 14,742 | -18% | 1 | 1 | 0% | 2,641 | 2,803 | +6% | 0 | 0 | — |
case-16 | fail→pass | 13,223 | 7,325 | -45% | 1 | 1 | 0% | 2,170 | 1,875 | -14% | 0 | 0 | — |
case-17 | fail→fail | 14,876 | 11,463 | -23% | 1 | 1 | 0% | 2,194 | 2,260 | +3% | 0 | 0 | — |
case-18 | pass→pass | 12,526 | 2,371 | -81% | 1 | 1 | 0% | 1,889 | 977 | -48% | 0 | 0 | — |
case-19 | fail→pass | 11,994 | 2,007 | -83% | 1 | 1 | 0% | 1,823 | 964 | -47% | 0 | 0 | — |
case-20 | pass→pass | 14,676 | 15,622 | +6% | 1 | 1 | 0% | 2,610 | 3,368 | +29% | 0 | 0 | — |
case-21 | pass→pass | 15,318 | 17,751 | +16% | 1 | 1 | 0% | 2,470 | 3,340 | +35% | 0 | 0 | — |
case-22 | pass→pass | 16,402 | 12,118 | -26% | 1 | 1 | 0% | 2,873 | 2,783 | -3% | 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, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +32 percentage points is the difference between those two pass rates over the 21 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.