Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when auditing account settings pages, privacy dashboards, or API routes to verify that a complete data deletion path exists for users.
.claude/skills/thedaviddias-right-to-erasure/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -9% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 28% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 33% | 0% |
| case-22 | ✓→✓ | ＝ Same ✓ | 38% | 0% |
GDPR Article 17 gives EU residents the right to have their personal data erased when it is no longer necessary for the purpose it was collected, or when they withdraw consent. Failing to honour this right can result in regulatory fines of up to €20 million or 4% of global annual turnover. Providing a clear deletion flow also builds trust with users who value control over their data.
Check whether this application provides a user-facing data deletion mechanism that clears all client-side storage and triggers a server-side deletion request.
Implement a data deletion flow that clears localStorage, sessionStorage, IndexedDB, and cookies, then sends a deletion request to the server and shows confirmation to the user.
Explain GDPR Article 17 (right to erasure) and what a compliant data deletion flow must cover, including both client-side and server-side data.
Review account settings, privacy pages, and API routes for a data deletion endpoint. Flag missing client-side storage clearance, absent confirmation UI, or missing server-side deletion calls.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/privacy/right-to-erasure
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | pass→pass | 11,943 | 8,489 | -29% | 1 | 1 | 0% | 2,291 | 2,091 | -9% | 0 | 0 | — |
case-20 | pass→pass | 12,710 | 14,108 | +11% | 1 | 1 | 0% | 2,359 | 3,030 | +28% | 0 | 0 | — |
case-21 | pass→pass | 13,448 | 14,110 | +5% | 1 | 1 | 0% | 2,248 | 2,992 | +33% | 0 | 0 | — |
case-22 | pass→pass | 13,644 | 15,260 | +12% | 1 | 1 | 0% | 2,209 | 3,046 | +38% | 0 | 0 | — |
case-01 | pass→pass | 13,815 | 7,014 | -49% | 1 | 1 | 0% | 2,264 | 1,677 | -26% | 0 | 0 | — |
case-03 | pass→pass | 9,308 | 4,599 | -51% | 1 | 1 | 0% | 1,573 | 1,184 | -25% | 0 | 0 | — |
case-04 | pass→pass | 5,653 | 1,883 | -67% | 1 | 1 | 0% | 999 | 699 | -30% | 0 | 0 | — |
case-05 | pass→pass | 8,526 | 7,312 | -14% | 1 | 1 | 0% | 1,710 | 1,771 | +4% | 0 | 0 | — |
case-06 | pass→pass | 5,926 | 6,346 | +7% | 1 | 1 | 0% | 1,058 | 1,564 | +48% | 0 | 0 | — |
case-07 | pass→pass | 16,250 | 11,283 | -31% | 1 | 1 | 0% | 3,165 | 2,711 | -14% | 0 | 0 | — |
case-08 | pass→pass | 13,228 | 9,652 | -27% | 1 | 1 | 0% | 2,301 | 2,197 | -5% | 0 | 0 | — |
case-09 | pass→pass | 13,795 | 10,782 | -22% | 1 | 1 | 0% | 2,345 | 2,328 | -1% | 0 | 0 | — |
case-10 | pass→pass | 10,651 | 6,133 | -42% | 1 | 1 | 0% | 2,017 | 1,505 | -25% | 0 | 0 | — |
case-11 | pass→pass | 10,224 | 8,110 | -21% | 1 | 1 | 0% | 1,726 | 1,808 | +5% | 0 | 0 | — |
case-12 | pass→pass | 11,565 | 7,652 | -34% | 1 | 1 | 0% | 2,156 | 1,961 | -9% | 0 | 0 | — |
case-13 | pass→pass | 4,026 | 2,861 | -29% | 1 | 1 | 0% | 723 | 822 | +14% | 0 | 0 | — |
case-14 | pass→pass | 6,910 | 5,383 | -22% | 1 | 1 | 0% | 1,142 | 1,373 | +20% | 0 | 0 | — |
case-15 | pass→pass | 11,359 | 9,588 | -16% | 1 | 1 | 0% | 1,910 | 2,109 | +10% | 0 | 0 | — |
case-16 | pass→pass | 11,433 | 10,506 | -8% | 1 | 1 | 0% | 1,873 | 2,202 | +18% | 0 | 0 | — |
case-17 | pass→pass | 5,366 | 3,250 | -39% | 1 | 1 | 0% | 935 | 888 | -5% | 0 | 0 | — |
case-18 | pass→pass | 13,044 | 8,512 | -35% | 1 | 1 | 0% | 2,387 | 1,971 | -17% | 0 | 0 | — |
case-19 | fail→pass | 14,243 | 11,001 | -23% | 1 | 1 | 0% | 2,424 | 2,329 | -4% | 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.