Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to create or optimize a cookie policy page. Also use when the user mentions "cookie policy," "cookies," "cookie consent," "GDPR cookies," "cookie banner," "cookie notice," "tracking cookies," or "cookie settings." For legal overview, use legal-page-generator.
.claude/skills/sediman-agent-cookie-policy-page-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 6% | 0% |
| case-14 | ✓→✓ | ＝ Same ✓ | -32% | 0% |
| case-15 | ✓→✓ | ＝ Same ✓ | 7% | 0% |
Guides cookie policy page content for GDPR and transparency. Often separate from Privacy Policy.
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:
| Element | Purpose | |---------|---------| | What are cookies | Brief explanation | | Types used | Essential, analytics, marketing | | Purpose | Why each type; duration | | Third parties | Who sets cookies (GA, ads) | | Management | How to opt out; preferences | | Updates | When policy changes |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 10,898 | 9,242 | -15% | 1 | 1 | 0% | 1,629 | 1,730 | +6% | 0 | 0 | — |
case-14 | pass→pass | 19,636 | 10,539 | -46% | 1 | 1 | 0% | 2,867 | 1,946 | -32% | 0 | 0 | — |
case-15 | pass→pass | 11,928 | 10,938 | -8% | 1 | 1 | 0% | 1,842 | 1,962 | +7% | 0 | 0 | — |
case-01 | pass→pass | 15,014 | 12,700 | -15% | 1 | 1 | 0% | 2,659 | 2,362 | -11% | 0 | 0 | — |
case-02 | pass→pass | 15,626 | 14,216 | -9% | 1 | 1 | 0% | 2,354 | 2,590 | +10% | 0 | 0 | — |
case-03 | pass→pass | 12,198 | 11,733 | -4% | 1 | 1 | 0% | 1,920 | 2,101 | +9% | 0 | 0 | — |
case-04 | fail→pass | 13,727 | 10,484 | -24% | 1 | 1 | 0% | 2,192 | 2,045 | -7% | 0 | 0 | — |
case-06 | pass→pass | 12,830 | 10,221 | -20% | 1 | 1 | 0% | 1,969 | 1,944 | -1% | 0 | 0 | — |
case-07 | pass→pass | 12,966 | 11,103 | -14% | 1 | 1 | 0% | 1,934 | 1,881 | -3% | 0 | 0 | — |
case-08 | pass→pass | 20,725 | 13,111 | -37% | 1 | 1 | 0% | 3,039 | 2,487 | -18% | 0 | 0 | — |
case-09 | fail→fail | 10,524 | 10,469 | -1% | 1 | 1 | 0% | 1,805 | 2,000 | +11% | 0 | 0 | — |
case-20 | pass→pass | 20,499 | 12,259 | -40% | 1 | 1 | 0% | 3,233 | 2,265 | -30% | 0 | 0 | — |
case-10 | pass→pass | 14,189 | 10,072 | -29% | 1 | 1 | 0% | 2,033 | 1,742 | -14% | 0 | 0 | — |
case-11 | fail→pass | 12,740 | 11,346 | -11% | 1 | 1 | 0% | 1,954 | 1,991 | +2% | 0 | 0 | — |
case-12 | pass→pass | 7,744 | 7,526 | -3% | 1 | 1 | 0% | 1,199 | 1,477 | +23% | 0 | 0 | — |
case-13 | pass→pass | 16,211 | 15,313 | -6% | 1 | 1 | 0% | 2,482 | 2,533 | +2% | 0 | 0 | — |
case-16 | pass→pass | 11,179 | 9,239 | -17% | 1 | 1 | 0% | 1,730 | 1,701 | -2% | 0 | 0 | — |
case-17 | fail→fail | 13,445 | 13,355 | -1% | 1 | 1 | 0% | 2,105 | 2,544 | +21% | 0 | 0 | — |
case-18 | pass→pass | 19,654 | 14,160 | -28% | 1 | 1 | 0% | 3,136 | 2,620 | -16% | 0 | 0 | — |
case-19 | pass→pass | 12,497 | 10,100 | -19% | 1 | 1 | 0% | 1,895 | 1,974 | +4% | 0 | 0 | — |
case-21 | pass→pass | 18,571 | 17,390 | -6% | 1 | 1 | 0% | 3,236 | 3,181 | -2% | 0 | 0 | — |
case-22 | pass→pass | 18,213 | 13,083 | -28% | 1 | 1 | 0% | 3,057 | 2,608 | -15% | 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 +9 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.