Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert in secure backend coding practices specializing in input validation, authentication, and API security. Use PROACTIVELY for backend security implementations or security code reviews.
.claude/skills/dokhacgiakhoa-backend-security-coder/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-08 | ✓→✓ | = Same ✓ | 33% | 0% |
| case-18 | ✓→✓ | = Same ✓ | 16% | 0% |
| case-19 | ✓→✓ | = Same ✓ | 97% | 0% |
| case-01 | ✗→✗ | = Same ✗ | 15% | 0% |
resources/implementation-playbook.md.You are a backend security coding expert specializing in secure development practices, vulnerability prevention, and secure architecture implementation.
Expert backend security developer with comprehensive knowledge of secure coding practices, vulnerability prevention, and defensive programming techniques. Masters input validation, authentication systems, API security, database protection, and secure error handling. Specializes in building security-first backend applications that resist common attack vectors.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,414 | 28,479 | +22% | 1 | 1 | 0% | 4,113 | 4,743 | +15% | 0 | 0 | — |
case-02 | fail→fail | 20,762 | 15,744 | -24% | 1 | 1 | 0% | 4,119 | 4,067 | -1% | 0 | 0 | — |
case-03 | fail→fail | 17,735 | 15,462 | -13% | 1 | 1 | 0% | 3,588 | 3,929 | +10% | 0 | 0 | — |
case-04 | fail→pass | 16,524 | 17,081 | +3% | 1 | 1 | 0% | 2,982 | 4,028 | +35% | 0 | 0 | — |
case-05 | fail→fail | 46,367 | 26,364 | -43% | 1 | 1 | 0% | 7,758 | 5,813 | -25% | 0 | 0 | — |
case-06 | fail→fail | 21,851 | 21,926 | +0% | 1 | 1 | 0% | 3,733 | 4,520 | +21% | 0 | 0 | — |
case-07 | fail→fail | 19,222 | 18,953 | -1% | 1 | 1 | 0% | 4,211 | 4,571 | +9% | 0 | 0 | — |
case-08 | pass→pass | 11,249 | 12,580 | +12% | 1 | 1 | 0% | 2,489 | 3,318 | +33% | 0 | 0 | — |
case-09 | fail→fail | 12,217 | 16,063 | +31% | 1 | 1 | 0% | 2,490 | 3,296 | +32% | 0 | 0 | — |
case-10 | fail→fail | 32,278 | 28,952 | -10% | 1 | 1 | 0% | 5,471 | 7,131 | +30% | 0 | 0 | — |
case-11 | fail→fail | 19,419 | 25,628 | +32% | 1 | 1 | 0% | 3,982 | 4,742 | +19% | 0 | 0 | — |
case-12 | fail→fail | 14,759 | 28,768 | +95% | 1 | 1 | 0% | 3,197 | 5,170 | +62% | 0 | 0 | — |
case-13 | fail→fail | 17,989 | 24,904 | +38% | 1 | 1 | 0% | 3,944 | 5,959 | +51% | 0 | 0 | — |
case-14 | fail→fail | 17,763 | 19,635 | +11% | 1 | 1 | 0% | 3,456 | 4,305 | +25% | 0 | 0 | — |
case-15 | fail→fail | 13,696 | 19,274 | +41% | 1 | 1 | 0% | 2,523 | 3,419 | +36% | 0 | 0 | — |
case-16 | fail→fail | 20,619 | 23,404 | +14% | 1 | 1 | 0% | 4,211 | 5,249 | +25% | 0 | 0 | — |
case-17 | fail→fail | 17,187 | 23,683 | +38% | 1 | 1 | 0% | 3,153 | 4,298 | +36% | 0 | 0 | — |
case-18 | pass→pass | 21,391 | 28,829 | +35% | 1 | 1 | 0% | 4,454 | 5,181 | +16% | 0 | 0 | — |
case-19 | pass→pass | 10,142 | 13,932 | +37% | 1 | 1 | 0% | 1,497 | 2,954 | +97% | 0 | 0 | — |
case-20 | fail→fail | 14,242 | 12,861 | -10% | 1 | 1 | 0% | 2,443 | 2,912 | +19% | 0 | 0 | — |
case-21 | fail→fail | 14,166 | 17,619 | +24% | 1 | 1 | 0% | 2,840 | 4,256 | +50% | 0 | 0 | — |
case-22 | fail→fail | 16,106 | 33,361 | +107% | 1 | 1 | 0% | 2,861 | 2,817 | -2% | 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.