Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to create, optimize, or structure legal pages (Privacy, Terms, etc.). Also use when the user mentions "privacy policy," "terms of service," "legal pages," "cookie policy," "terms and conditions," "legal footer," "legal section," "compliance pages," or "legal requirements." For Privacy Policy content, use privacy-page-generator. For Terms of Service, use terms-page-generator. For Cookie Policy, use cookie-policy-page-generator.
.claude/skills/sediman-agent-legal-page-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 18% | 0% |
Guides legal page content, structure, and SEO handling.
When invoking: On first use, if helpful, open with 1–2 sentences on what this skill covers and why it matters, then provide the main output. On subsequent use or when the user asks to skip, go directly to the main output.
Identify:
| Page | Purpose | |------|---------| | Privacy Policy | Data collection, use, sharing, rights | | Terms of Service | User agreement, limitations, termination | | Cookie Policy | Cookie types, consent, management | | Acceptable Use | Prohibited uses, enforcement | | Refund Policy | Refund conditions (e-commerce) |
| Approach | When | |----------|------| | Index | If you want legal pages discoverable (rare) | | Noindex | Common for legal; reduces low-value indexed pages | | Canonical | If multiple versions (e.g., by region) |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | fail→pass | 19,361 | 17,120 | -12% | 1 | 1 | 0% | 3,105 | 3,339 | +8% | 0 | 0 | — |
case-06 | fail→fail | 15,033 | 14,758 | -2% | 1 | 1 | 0% | 2,427 | 2,841 | +17% | 0 | 0 | — |
case-07 | pass→pass | 15,061 | 11,705 | -22% | 1 | 1 | 0% | 2,464 | 2,458 | -0% | 0 | 0 | — |
case-04 | pass→pass | 19,910 | 20,327 | +2% | 1 | 1 | 0% | 3,288 | 4,051 | +23% | 0 | 0 | — |
case-01 | fail→pass | 21,410 | 18,668 | -13% | 1 | 1 | 0% | 3,321 | 3,735 | +12% | 0 | 0 | — |
case-02 | pass→pass | 33,093 | 21,174 | -36% | 1 | 1 | 0% | 5,662 | 4,046 | -29% | 0 | 0 | — |
case-03 | fail→fail | 20,120 | 14,416 | -28% | 1 | 1 | 0% | 3,274 | 2,877 | -12% | 0 | 0 | — |
case-08 | pass→pass | 19,895 | 14,436 | -27% | 1 | 1 | 0% | 3,128 | 2,746 | -12% | 0 | 0 | — |
case-09 | fail→pass | 16,461 | 13,720 | -17% | 1 | 1 | 0% | 2,652 | 2,825 | +7% | 0 | 0 | — |
case-10 | fail→pass | 17,916 | 16,358 | -9% | 1 | 1 | 0% | 2,846 | 3,146 | +11% | 0 | 0 | — |
case-11 | fail→pass | 16,908 | 16,453 | -3% | 1 | 1 | 0% | 2,811 | 3,320 | +18% | 0 | 0 | — |
case-12 | pass→pass | 20,075 | 14,652 | -27% | 1 | 1 | 0% | 3,333 | 3,033 | -9% | 0 | 0 | — |
case-13 | fail→pass | 20,121 | 17,310 | -14% | 1 | 1 | 0% | 3,406 | 3,477 | +2% | 0 | 0 | — |
case-14 | pass→pass | 14,917 | 13,435 | -10% | 1 | 1 | 0% | 2,637 | 2,675 | +1% | 0 | 0 | — |
case-15 | pass→pass | 17,592 | 13,752 | -22% | 1 | 1 | 0% | 2,886 | 2,754 | -5% | 0 | 0 | — |
case-16 | pass→pass | 16,828 | 13,838 | -18% | 1 | 1 | 0% | 2,569 | 2,567 | -0% | 0 | 0 | — |
case-17 | pass→fail | 14,028 | 14,155 | +1% | 1 | 1 | 0% | 2,304 | 2,957 | +28% | 0 | 0 | — |
case-18 | fail→pass | 17,110 | 13,774 | -19% | 1 | 1 | 0% | 2,734 | 2,793 | +2% | 0 | 0 | — |
case-19 | fail→fail | 15,812 | 15,265 | -3% | 1 | 1 | 0% | 2,641 | 3,079 | +17% | 0 | 0 | — |
case-20 | fail→fail | 34,851 | 25,478 | -27% | 1 | 1 | 0% | 6,175 | 4,989 | -19% | 0 | 0 | — |
case-21 | fail→pass | 15,849 | 15,701 | -1% | 1 | 1 | 0% | 2,695 | 3,342 | +24% | 0 | 0 | — |
case-22 | fail→pass | 26,478 | 18,884 | -29% | 1 | 1 | 0% | 4,669 | 4,048 | -13% | 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 +36 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.