Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Audit what your public online presence quietly reveals — and tighten it — before a stranger, employer, or scammer uses it. Use when asked what does my online presence reveal, audit my privacy, what can people find out about me, or clean up my social media. Produces a review of what's exposed across profiles and posts (location, routines, identifiers, security-question answers), the specific risks each creates, prioritized fixes (settings + what to remove/stop posting), and habits to prevent futu
.claude/skills/mohitagw15856-oversharing-audit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | 90% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 22% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 39% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 45% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 14% | 0% |
The pieces you post look harmless one at a time — a pet's name, a gym check-in, a birthday, a boarding pass photo — but together they hand strangers your routines, your location, and the answers to your security questions. This audits what your public presence actually reveals, ranks the real risks, and tightens things up without telling you to delete your whole life.
Ask for these if not provided:
What's exposed (as an outsider sees it)
Fix first (high risk): settings to change · posts/fields to remove or lock]. Then: medium-risk cleanup]. Safer habits: post trips after · no home geotag · crop identifiers · keep security answers private]. Keep: what's fine to leave].
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 24,146 | 30,310 | +26% | 1 | 1 | 0% | 3,152 | 4,375 | +39% | 0 | 0 | — |
case-05 | pass→pass | 16,058 | 16,589 | +3% | 1 | 1 | 0% | 2,015 | 2,924 | +45% | 0 | 0 | — |
case-03 | pass→fail | 26,339 | 25,133 | -5% | 1 | 1 | 0% | 2,873 | 3,497 | +22% | 0 | 0 | — |
case-01 | pass→pass | 24,934 | 41,968 | +68% | 1 | 1 | 0% | 2,698 | 3,082 | +14% | 0 | 0 | — |
case-02 | pass→pass | 32,791 | 21,850 | -33% | 1 | 1 | 0% | 3,062 | 3,442 | +12% | 0 | 0 | — |
case-06 | pass→pass | 22,204 | 21,895 | -1% | 1 | 1 | 0% | 3,025 | 3,802 | +26% | 0 | 0 | — |
case-07 | pass→pass | 20,214 | 19,548 | -3% | 1 | 1 | 0% | 1,995 | 2,893 | +45% | 0 | 0 | — |
case-08 | pass→pass | 17,748 | 21,858 | +23% | 1 | 1 | 0% | 1,957 | 3,116 | +59% | 0 | 0 | — |
case-09 | pass→pass | 18,006 | 11,626 | -35% | 1 | 1 | 0% | 1,834 | 2,624 | +43% | 0 | 0 | — |
case-10 | pass→pass | 19,425 | 25,382 | +31% | 1 | 1 | 0% | 1,869 | 2,448 | +31% | 0 | 0 | — |
case-11 | pass→pass | 18,071 | 19,658 | +9% | 1 | 1 | 0% | 1,870 | 2,811 | +50% | 0 | 0 | — |
case-12 | pass→pass | 18,410 | 15,109 | -18% | 1 | 1 | 0% | 2,087 | 2,881 | +38% | 0 | 0 | — |
case-13 | pass→pass | 21,875 | 13,707 | -37% | 1 | 1 | 0% | 1,976 | 2,746 | +39% | 0 | 0 | — |
case-14 | pass→pass | 21,095 | 13,079 | -38% | 1 | 1 | 0% | 2,033 | 2,606 | +28% | 0 | 0 | — |
case-15 | pass→pass | 22,798 | 16,170 | -29% | 1 | 1 | 0% | 2,095 | 2,901 | +38% | 0 | 0 | — |
case-16 | pass→pass | 17,328 | 14,004 | -19% | 1 | 1 | 0% | 1,777 | 3,124 | +76% | 0 | 0 | — |
case-17 | pass→pass | 22,004 | 23,539 | +7% | 1 | 1 | 0% | 2,145 | 3,221 | +50% | 0 | 0 | — |
case-18 | pass→pass | 20,324 | 28,608 | +41% | 1 | 1 | 0% | 2,130 | 2,704 | +27% | 0 | 0 | — |
case-19 | pass→pass | 24,274 | 38,706 | +59% | 1 | 1 | 0% | 1,938 | 3,100 | +60% | 0 | 0 | — |
case-20 | pass→pass | 25,703 | 15,508 | -40% | 1 | 1 | 0% | 2,052 | 3,174 | +55% | 0 | 0 | — |
case-21 | fail→pass | 13,461 | 15,013 | +12% | 1 | 1 | 0% | 1,657 | 3,153 | +90% | 0 | 0 | — |
case-22 | pass→pass | 15,613 | 16,094 | +3% | 1 | 1 | 0% | 1,929 | 2,925 | +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. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 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.