Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Check compliance with OWASP Top 10 security risks and best practices. Use when performing comprehensive security audits. Trigger with 'check OWASP compliance', 'audit web security', or 'validate OWASP'.
.claude/skills/jeremylongshore-checking-owasp-compliance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 885% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 75% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-22 | ✓→✗ | ▼ Worse | 22% | 0% |
Assess web applications against the OWASP Top 10, identifying injection flaws, broken authentication, sensitive data exposure, XXE, access control gaps, and security misconfigurations.
This skill empowers Claude to assess your project's adherence to the OWASP Top 10 (2021) security guidelines. It automates the process of identifying potential vulnerabilities related to common web application security risks, providing actionable insights to improve your application's security posture.
This skill activates when you need to:
User request: "Check OWASP compliance for SQL injection vulnerabilities."
The skill will:
User request: "/owasp"
The skill will:
This skill can be integrated with other plugins to automate vulnerability remediation or generate comprehensive security reports. For example, it can be used in conjunction with a code modification plugin to automatically apply recommended fixes for identified vulnerabilities.
If security scanning fails:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 11,752 | 8,173 | -30% | 1 | 1 | 0% | 1,440 | 1,680 | +17% | 0 | 0 | — |
case-02 | fail→fail | 11,016 | 8,534 | -23% | 1 | 1 | 0% | 1,125 | 1,606 | +43% | 0 | 0 | — |
case-03 | fail→pass | 3,831 | 23,614 | +516% | 1 | 1 | 0% | 603 | 5,942 | +885% | 0 | 0 | — |
case-04 | fail→fail | 5,801 | 5,118 | -12% | 1 | 1 | 0% | 1,098 | 1,070 | -3% | 0 | 0 | — |
case-05 | fail→fail | 25,398 | 18,751 | -26% | 1 | 1 | 0% | 4,500 | 4,170 | -7% | 0 | 0 | — |
case-06 | pass→pass | 14,711 | 12,312 | -16% | 1 | 1 | 0% | 2,690 | 3,089 | +15% | 0 | 0 | — |
case-07 | pass→pass | 16,596 | 15,728 | -5% | 1 | 1 | 0% | 2,596 | 2,189 | -16% | 0 | 0 | — |
case-08 | pass→pass | 11,329 | 11,092 | -2% | 1 | 1 | 0% | 1,905 | 2,858 | +50% | 0 | 0 | — |
case-09 | fail→pass | 9,606 | 10,898 | +13% | 1 | 1 | 0% | 1,849 | 2,839 | +54% | 0 | 0 | — |
case-10 | pass→pass | 13,799 | 14,807 | +7% | 1 | 1 | 0% | 2,298 | 3,232 | +41% | 0 | 0 | — |
case-11 | pass→pass | 11,753 | 9,742 | -17% | 1 | 1 | 0% | 2,030 | 2,670 | +32% | 0 | 0 | — |
case-12 | pass→pass | 12,866 | 10,082 | -22% | 1 | 1 | 0% | 2,430 | 2,608 | +7% | 0 | 0 | — |
case-13 | pass→pass | 13,729 | 13,692 | -0% | 1 | 1 | 0% | 2,574 | 3,358 | +30% | 0 | 0 | — |
case-14 | pass→pass | 14,377 | 12,254 | -15% | 1 | 1 | 0% | 2,127 | 2,909 | +37% | 0 | 0 | — |
case-15 | pass→pass | 11,354 | 12,122 | +7% | 1 | 1 | 0% | 1,930 | 2,888 | +50% | 0 | 0 | — |
case-16 | pass→pass | 11,144 | 12,419 | +11% | 1 | 1 | 0% | 1,875 | 2,879 | +54% | 0 | 0 | — |
case-17 | pass→pass | 14,404 | 11,183 | -22% | 1 | 1 | 0% | 2,516 | 2,716 | +8% | 0 | 0 | — |
case-18 | fail→pass | 11,556 | 16,044 | +39% | 1 | 1 | 0% | 2,146 | 3,745 | +75% | 0 | 0 | — |
case-19 | fail→pass | 13,527 | 14,877 | +10% | 1 | 1 | 0% | 2,192 | 3,263 | +49% | 0 | 0 | — |
case-20 | pass→pass | 12,396 | 10,645 | -14% | 1 | 1 | 0% | 2,094 | 2,721 | +30% | 0 | 0 | — |
case-21 | pass→pass | 15,309 | 12,610 | -18% | 1 | 1 | 0% | 2,639 | 2,991 | +13% | 0 | 0 | — |
case-22 | pass→fail | 16,472 | 15,907 | -3% | 1 | 1 | 0% | 2,988 | 3,659 | +22% | 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 +14 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.