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 Terms of Service page. Also use when the user mentions "terms of service," "terms and conditions," "terms of use," "user agreement," "ToS," "legal terms," "service agreement," or "terms page." For legal overview page, use legal-page-generator.
.claude/skills/sediman-agent-terms-page-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | -6% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 15% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -19% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -20% | 0% |
Guides Terms of Service page content, structure, and compliance.
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:
| Section | Content | |---------|---------| | Acceptance | How agreement is formed (signup, use) | | Service description | What you provide | | User obligations | Acceptable use, account security | | Intellectual property | Who owns what | | Payment | If applicable; billing, refunds | | Liability | Limitations, disclaimers | | Termination | When and how accounts end | | Governing law | Jurisdiction, dispute resolution | | Changes | How you'll notify of updates | | Contact | How to reach you about terms |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 16,164 | 13,939 | -14% | 1 | 1 | 0% | 2,698 | 2,531 | -6% | 0 | 0 | — |
case-07 | pass→pass | 8,683 | 7,774 | -10% | 1 | 1 | 0% | 1,405 | 1,612 | +15% | 0 | 0 | — |
case-01 | fail→pass | 22,918 | 17,959 | -22% | 1 | 1 | 0% | 3,867 | 3,282 | -15% | 0 | 0 | — |
case-02 | fail→fail | 23,344 | 17,714 | -24% | 1 | 1 | 0% | 3,723 | 3,220 | -14% | 0 | 0 | — |
case-03 | pass→pass | 21,264 | 14,153 | -33% | 1 | 1 | 0% | 3,307 | 2,686 | -19% | 0 | 0 | — |
case-04 | pass→pass | 20,797 | 13,441 | -35% | 1 | 1 | 0% | 3,078 | 2,456 | -20% | 0 | 0 | — |
case-05 | pass→pass | 18,491 | 11,241 | -39% | 1 | 1 | 0% | 2,801 | 2,136 | -24% | 0 | 0 | — |
case-06 | pass→pass | 17,154 | 12,743 | -26% | 1 | 1 | 0% | 2,602 | 2,439 | -6% | 0 | 0 | — |
case-08 | pass→pass | 14,347 | 11,105 | -23% | 1 | 1 | 0% | 2,290 | 2,160 | -6% | 0 | 0 | — |
case-09 | pass→pass | 13,523 | 10,517 | -22% | 1 | 1 | 0% | 2,068 | 2,032 | -2% | 0 | 0 | — |
case-10 | pass→pass | 16,716 | 10,929 | -35% | 1 | 1 | 0% | 2,490 | 2,054 | -18% | 0 | 0 | — |
case-11 | pass→pass | 12,516 | 10,268 | -18% | 1 | 1 | 0% | 1,786 | 1,975 | +11% | 0 | 0 | — |
case-12 | fail→fail | 13,266 | 13,322 | +0% | 1 | 1 | 0% | 1,933 | 2,379 | +23% | 0 | 0 | — |
case-13 | pass→pass | 14,398 | 12,890 | -10% | 1 | 1 | 0% | 2,301 | 2,286 | -1% | 0 | 0 | — |
case-14 | fail→fail | 9,187 | 8,547 | -7% | 1 | 1 | 0% | 1,430 | 1,715 | +20% | 0 | 0 | — |
case-15 | pass→pass | 16,861 | 10,115 | -40% | 1 | 1 | 0% | 2,355 | 1,873 | -20% | 0 | 0 | — |
case-16 | pass→pass | 14,383 | 9,588 | -33% | 1 | 1 | 0% | 1,944 | 1,852 | -5% | 0 | 0 | — |
case-17 | pass→pass | 16,818 | 12,157 | -28% | 1 | 1 | 0% | 2,418 | 2,261 | -6% | 0 | 0 | — |
case-18 | pass→pass | 16,935 | 12,948 | -24% | 1 | 1 | 0% | 2,511 | 2,468 | -2% | 0 | 0 | — |
case-19 | pass→pass | 21,725 | 14,136 | -35% | 1 | 1 | 0% | 3,351 | 2,495 | -26% | 0 | 0 | — |
case-21 | pass→pass | 17,771 | 15,700 | -12% | 1 | 1 | 0% | 2,919 | 2,970 | +2% | 0 | 0 | — |
case-22 | pass→pass | 27,948 | 14,715 | -47% | 1 | 1 | 0% | 5,263 | 3,245 | -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 +5 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.