Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Get your personal info off people-search and data-broker sites — a prioritized opt-out plan that targets the sites that matter and keeps them from reappearing. Use when asked to remove my info from the internet, opt out of data brokers, my address/phone is on people-search sites, or reduce my digital footprint. Produces a prioritized target list (the high-traffic brokers first), the opt-out method for each, a suppression-at-source plan so data stops flowing back, a recheck cadence, and safe-hand
.claude/skills/mohitagw15856-data-broker-removal/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-07 | ✓→✓ | = Same ✓ | 31% | 0% |
| case-08 | ✓→✓ | = Same ✓ | 81% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 9% | 0% |
Your home address, phone, relatives, and past addresses are listed on dozens of people-search sites — scraped and resold. You can't clear all of them forever, but you can remove the highest-impact ones and slow the reappearance. This gives a prioritized opt-out plan: which brokers to hit first, how to opt out of each, and how to reduce the sources feeding them.
Ask for these if not provided:
Map it: self-search results → what's exposed.
Remove (priority order)
… (highest-traffic first)
Suppress the source: public-record/marketing/account settings]. Recheck: every 1–3 months] — re-scan and re-remove. Cautions: submit only required info · avoid shady "removal" services · safety note if relevant].
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | pass→pass | 20,393 | 19,726 | -3% | 1 | 1 | 0% | 2,403 | 3,152 | +31% | 0 | 0 | — |
case-08 | pass→pass | 17,424 | 21,419 | +23% | 1 | 1 | 0% | 1,618 | 2,923 | +81% | 0 | 0 | — |
case-01 | fail→fail | 29,648 | 29,364 | -1% | 1 | 1 | 0% | 3,862 | 4,204 | +9% | 0 | 0 | — |
case-02 | fail→fail | 27,159 | 35,424 | +30% | 1 | 1 | 0% | 3,426 | 4,434 | +29% | 0 | 0 | — |
case-03 | fail→pass | 37,217 | 27,607 | -26% | 1 | 1 | 0% | 5,211 | 4,030 | -23% | 0 | 0 | — |
case-04 | pass→pass | 25,867 | 25,556 | -1% | 1 | 1 | 0% | 3,276 | 3,565 | +9% | 0 | 0 | — |
case-05 | pass→pass | 22,070 | 27,282 | +24% | 1 | 1 | 0% | 2,593 | 3,951 | +52% | 0 | 0 | — |
case-06 | pass→pass | 18,612 | 20,855 | +12% | 1 | 1 | 0% | 2,386 | 3,463 | +45% | 0 | 0 | — |
case-09 | pass→pass | 17,600 | 21,453 | +22% | 1 | 1 | 0% | 1,987 | 3,111 | +57% | 0 | 0 | — |
case-10 | fail→fail | 20,595 | 23,384 | +14% | 1 | 1 | 0% | 2,061 | 3,088 | +50% | 0 | 0 | — |
case-11 | pass→pass | 18,125 | 20,258 | +12% | 1 | 1 | 0% | 1,693 | 2,923 | +73% | 0 | 0 | — |
case-12 | pass→pass | 33,645 | 26,552 | -21% | 1 | 1 | 0% | 2,539 | 3,900 | +54% | 0 | 0 | — |
case-13 | pass→pass | 23,675 | 28,981 | +22% | 1 | 1 | 0% | 2,342 | 4,006 | +71% | 0 | 0 | — |
case-14 | fail→pass | 21,151 | 22,448 | +6% | 1 | 1 | 0% | 2,289 | 3,232 | +41% | 0 | 0 | — |
case-15 | pass→pass | 25,096 | 23,881 | -5% | 1 | 1 | 0% | 2,354 | 3,797 | +61% | 0 | 0 | — |
case-16 | pass→pass | 38,970 | 33,378 | -14% | 1 | 1 | 0% | 2,387 | 4,073 | +71% | 0 | 0 | — |
case-17 | pass→pass | 16,327 | 16,962 | +4% | 1 | 1 | 0% | 1,659 | 2,329 | +40% | 0 | 0 | — |
case-18 | pass→pass | 18,548 | 22,053 | +19% | 1 | 1 | 0% | 1,616 | 3,404 | +111% | 0 | 0 | — |
case-19 | pass→pass | 12,812 | 15,841 | +24% | 1 | 1 | 0% | 1,846 | 2,616 | +42% | 0 | 0 | — |
case-20 | pass→pass | 32,585 | 23,075 | -29% | 1 | 1 | 0% | 2,131 | 3,118 | +46% | 0 | 0 | — |
case-21 | pass→pass | 34,006 | 25,622 | -25% | 1 | 1 | 0% | 3,497 | 3,885 | +11% | 0 | 0 | — |
case-22 | fail→fail | 32,010 | 30,044 | -6% | 1 | 1 | 0% | 4,186 | 4,404 | +5% | 0 | 0 | — |
case-23 | pass→pass | 19,001 | 33,768 | +78% | 1 | 1 | 0% | 2,497 | 3,800 | +52% | 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. 23 cases were attempted. The headline lift of +9 percentage points is the difference between those two pass rates over the 23 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.