Install any skill in seconds. Free to start, no credit card required.
Get Started Free →SecOps for application security, vulnerability management, compliance, and secure development. Use when implementing security controls, conducting security audits, responding to vulnerabilities, or meeting compliance requirements.
.claude/skills/borghei-senior-secops/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-15 | ✗→✓ | ▲ Improved | 45% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 96% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 86% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 62% | 0% |
The agent scans source code for security vulnerabilities (hardcoded secrets, SQL injection, XSS, command injection), assesses dependency CVEs across npm/Python/Go ecosystems, and verifies compliance against SOC 2, PCI-DSS, HIPAA, and GDPR frameworks.
Before the security pass, confirm these inputs. If any is unknown or vague, ASK — do not assume:
--framework; changes which controls are verified)--severity; changes the report and CI pass/fail)Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
| Tool | Purpose | Command | |------|---------|---------| | security_scanner.py | Scan source for secrets, injection, XSS, command injection, path traversal | python scripts/security_scanner.py <target> --severity high --json --output report.json | | vulnerability_assessor.py | Scan dependency manifests for known CVEs and compute risk score | python scripts/vulnerability_assessor.py <target> --severity critical | | compliance_checker.py | Verify SOC 2 / PCI-DSS / HIPAA / GDPR controls | python scripts/compliance_checker.py <target> --framework soc2 --json --output soc2.json |
Load the reference that matches the task — keep this file lean and pull detail on demand:
This skill covers:
This skill does NOT cover:
senior-devops skill for IaC hardening.senior-devops skill or use Trivy directly.| Skill | Integration | Data Flow | |-------|-------------|-----------| | senior-devops | Infrastructure hardening and CI/CD pipeline configuration | Security scan results feed into deployment gates; DevOps provides container and IaC scanning | | senior-backend | Secure coding patterns and input validation in server-side code | SecOps scanner findings drive backend remediation; backend applies parameterized queries and output encoding | | senior-qa | Security test cases and regression verification after patches | Vulnerability reports generate QA test cases; QA confirms fixes do not introduce regressions | | senior-architect | Threat modeling, defense-in-depth design, and zero-trust architecture | Compliance gaps inform architecture decisions; architect provides security design patterns | | code-reviewer | Security-focused code review and pre-merge analysis | Scanner findings prioritize review focus areas; reviewer enforces secure coding standards | | senior-fullstack | End-to-end security across frontend and API layers (XSS, CSRF, auth) | SecOps identifies frontend and API vulnerabilities; fullstack applies framework-level mitigations |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-06 | pass→pass | 14,092 | 8,135 | -42% | 1 | 1 | 0% | 2,406 | 2,791 | +16% | 0 | 0 | — |
case-15 | fail→pass | 6,494 | 2,299 | -65% | 1 | 1 | 0% | 1,195 | 1,735 | +45% | 0 | 0 | — |
case-01 | fail→fail | 9,501 | 9,931 | +5% | 1 | 1 | 0% | 1,082 | 2,349 | +117% | 0 | 0 | — |
case-02 | fail→fail | 28,147 | 34,874 | +24% | 1 | 1 | 0% | 4,399 | 7,015 | +59% | 0 | 0 | — |
case-03 | fail→fail | 6,779 | 25,304 | +273% | 1 | 1 | 0% | 1,174 | 5,956 | +407% | 0 | 0 | — |
case-04 | pass→pass | 6,772 | 8,850 | +31% | 1 | 1 | 0% | 667 | 2,150 | +222% | 0 | 0 | — |
case-05 | fail→pass | 15,003 | 26,723 | +78% | 1 | 1 | 0% | 2,860 | 5,600 | +96% | 0 | 0 | — |
case-07 | fail→fail | 7,980 | 6,607 | -17% | 1 | 1 | 0% | 891 | 1,829 | +105% | 0 | 0 | — |
case-08 | fail→pass | 15,753 | 5,759 | -63% | 1 | 1 | 0% | 2,403 | 2,352 | -2% | 0 | 0 | — |
case-09 | fail→fail | 9,243 | 7,346 | -21% | 1 | 1 | 0% | 1,069 | 2,133 | +100% | 0 | 0 | — |
case-10 | fail→pass | 5,796 | 13,379 | +131% | 1 | 1 | 0% | 971 | 1,807 | +86% | 0 | 0 | — |
case-11 | fail→pass | 8,006 | 5,260 | -34% | 1 | 1 | 0% | 1,355 | 2,201 | +62% | 0 | 0 | — |
case-12 | fail→pass | 10,228 | 2,373 | -77% | 1 | 1 | 0% | 1,367 | 1,835 | +34% | 0 | 0 | — |
case-13 | pass→pass | 3,444 | 5,767 | +67% | 1 | 1 | 0% | 509 | 1,967 | +286% | 0 | 0 | — |
case-14 | pass→pass | 4,322 | 3,055 | -29% | 1 | 1 | 0% | 621 | 1,787 | +188% | 0 | 0 | — |
case-16 | fail→pass | 21,800 | 3,694 | -83% | 1 | 1 | 0% | 1,962 | 2,023 | +3% | 0 | 0 | — |
case-17 | fail→pass | 7,482 | 3,757 | -50% | 1 | 1 | 0% | 1,295 | 1,893 | +46% | 0 | 0 | — |
case-18 | fail→pass | 9,343 | 2,152 | -77% | 1 | 1 | 0% | 1,454 | 1,776 | +22% | 0 | 0 | — |
case-19 | fail→pass | 13,184 | 2,471 | -81% | 1 | 1 | 0% | 2,086 | 1,820 | -13% | 0 | 0 | — |
case-20 | fail→pass | 9,181 | 4,861 | -47% | 1 | 1 | 0% | 1,299 | 1,941 | +49% | 0 | 0 | — |
case-21 | fail→pass | 7,404 | 4,589 | -38% | 1 | 1 | 0% | 1,206 | 2,149 | +78% | 0 | 0 | — |
case-22 | fail→pass | 8,754 | 3,942 | -55% | 1 | 1 | 0% | 1,304 | 1,771 | +36% | 0 | 0 | — |
case-23 | pass→pass | 10,489 | 2,881 | -73% | 1 | 1 | 0% | 1,619 | 1,803 | +11% | 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 +57 percentage points is the difference between those two pass rates over the 23 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.