Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to create, optimize, or structure Privacy Policy page. Also use when the user mentions "privacy policy," "privacy page," "data protection," "GDPR compliance," "privacy notice," "data privacy," "CCPA," "cookie policy," or "personal data." For legal overview page, use legal-page-generator.
.claude/skills/kostja94-privacy-page-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 67% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 96% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 202% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 108% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 65% | 0% |
Guides Privacy Policy page content, structure, compliance, and AI-specific disclosures.
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:
Structure with clear sub-sections:
What We Do NOT Collect (trust signal — include for all products):
Data You Submit (URLs, prompts, uploads, form inputs):
Generated Content (AI output, images, text):
Analytics Data:
Account Data (if applicable):
Payment Data (if applicable):
Table format with opt-out instructions:
| Cookie | Purpose | Duration | How to Opt Out | |--------|---------|----------|----------------| | _ga, _ga_* | Google Analytics — aggregate usage measurement | Up to 2 years | GA opt-out browser add-on or block third-party cookies | | [session_cookie] | Session management / fair-use enforcement | 24 hours / session | Clear browser data |
Also state:
Table format with privacy policy URLs:
| Provider | Purpose | Privacy Policy | |----------|---------|----------------| | Cloud provider] | Hosting / backend | URL] | | CDN provider] | Content delivery / routing | URL] | | AI model provider 1] | AI generation / processing | URL] | | AI model provider 2] | AI generation / processing | URL] | | Analytics provider] | Usage analytics | URL] |
Specify per category — avoid vague "as long as needed" language:
Structure by jurisdiction:
GDPR (EEA/UK):
CCPA (California):
How to Exercise:
Follow the model of leading AI platforms: append jurisdiction-specific modules. At minimum:
EEA/UK Supplement:
California Supplement:
Other supplements (China PIPL, Brazil LGPD, India DPDP) when the product has significant users in those jurisdictions.
These are mandatory for any product that uses AI models:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | fail→pass | 12,719 | 9,346 | -27% | 1 | 1 | 0% | 2,385 | 3,977 | +67% | 0 | 0 | — |
case-06 | fail→pass | 12,592 | 10,041 | -20% | 1 | 1 | 0% | 2,053 | 4,021 | +96% | 0 | 0 | — |
case-01 | fail→fail | 21,187 | 15,178 | -28% | 1 | 1 | 0% | 3,632 | 5,159 | +42% | 0 | 0 | — |
case-02 | fail→fail | 29,605 | 15,308 | -48% | 1 | 1 | 0% | 5,447 | 5,087 | -7% | 0 | 0 | — |
case-03 | fail→fail | 18,731 | 19,051 | +2% | 1 | 1 | 0% | 3,466 | 5,356 | +55% | 0 | 0 | — |
case-04 | pass→pass | 11,845 | 12,140 | +2% | 1 | 1 | 0% | 2,058 | 4,271 | +108% | 0 | 0 | — |
case-05 | pass→pass | 13,660 | 9,732 | -29% | 1 | 1 | 0% | 2,355 | 3,895 | +65% | 0 | 0 | — |
case-07 | pass→pass | 11,904 | 6,412 | -46% | 1 | 1 | 0% | 1,954 | 3,361 | +72% | 0 | 0 | — |
case-08 | pass→pass | 16,920 | 11,592 | -31% | 1 | 1 | 0% | 2,820 | 4,395 | +56% | 0 | 0 | — |
case-09 | pass→pass | 16,694 | 14,455 | -13% | 1 | 1 | 0% | 3,046 | 4,828 | +59% | 0 | 0 | — |
case-10 | pass→pass | 13,040 | 10,365 | -21% | 1 | 1 | 0% | 2,573 | 4,216 | +64% | 0 | 0 | — |
case-11 | pass→pass | 16,229 | 13,044 | -20% | 1 | 1 | 0% | 2,981 | 4,853 | +63% | 0 | 0 | — |
case-12 | pass→pass | 11,316 | 11,050 | -2% | 1 | 1 | 0% | 1,803 | 4,144 | +130% | 0 | 0 | — |
case-14 | pass→pass | 13,246 | 14,990 | +13% | 1 | 1 | 0% | 2,567 | 5,030 | +96% | 0 | 0 | — |
case-15 | pass→pass | 10,960 | 8,579 | -22% | 1 | 1 | 0% | 1,951 | 3,774 | +93% | 0 | 0 | — |
case-16 | pass→pass | 13,218 | 11,078 | -16% | 1 | 1 | 0% | 2,419 | 4,180 | +73% | 0 | 0 | — |
case-17 | pass→pass | 14,787 | 7,056 | -52% | 1 | 1 | 0% | 2,588 | 3,583 | +38% | 0 | 0 | — |
case-18 | fail→pass | 18,255 | 7,595 | -58% | 1 | 1 | 0% | 1,209 | 3,647 | +202% | 0 | 0 | — |
case-19 | pass→pass | 13,535 | 10,331 | -24% | 1 | 1 | 0% | 2,314 | 4,031 | +74% | 0 | 0 | — |
case-20 | fail→fail | 12,369 | 10,299 | -17% | 1 | 1 | 0% | 2,440 | 4,342 | +78% | 0 | 0 | — |
case-21 | pass→pass | 15,768 | 13,671 | -13% | 1 | 1 | 0% | 3,967 | 5,730 | +44% | 0 | 0 | — |
case-22 | pass→pass | 9,258 | 7,550 | -18% | 1 | 1 | 0% | 1,783 | 3,733 | +109% | 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 +14 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.