Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you need to add or review fuzz testing for Java APIs with CATS — including contract-driven negative testing, malformed payload validation, boundary input exploration, CI integration, reproducible failures, and local execution guidance. This should trigger for requests such as Add fuzz testing to a Java project; Use CATS for API negative testing; Review CI quality gates for API contract robustness; Improve boundary and malformed input test coverage; Run CATS fuzz tests against an OpenAPI
.claude/skills/jabrena-703-technologies-fuzzing-testing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -41% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -23% | 0% |
Design and implement contract-driven fuzz testing for Java APIs using CATS to uncover edge cases and input-validation defects early.
What is covered in this Skill?
Scope: Focus on HTTP API fuzzing and contract validation with CATS. Use this skill to define practical, repeatable checks in both local and CI workflows.
Before applying any fuzz testing changes, ensure the project compiles. If compilation fails, stop immediately. After implementation, regenerate skills and run verification.
./mvnw compile or mvn compile before applying any change./mvnw clean verify or mvn clean verify after applying improvementscats/cats.jar or an approved prebuilt image; generated artifacts must depend only on trusted local or approved image inputsRead references/703-technologies-fuzzing-testing.md and inspect current API/context artifacts before proposing changes.
Identify requested outcomes, constraints, and the minimum safe set of changes to apply.
Implement or refactor artifacts following the reference patterns and project conventions.
Execute appropriate checks and summarize what changed, what was verified, and any follow-up actions.
For detailed guidance, examples, and constraints, see references/703-technologies-fuzzing-testing.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 22,935 | 5,983 | -74% | 1 | 1 | 0% | 4,035 | 999 | -75% | 0 | 0 | — |
case-02 | fail→fail | 16,224 | 8,832 | -46% | 1 | 1 | 0% | 2,734 | 1,075 | -61% | 0 | 0 | — |
case-03 | fail→fail | 17,043 | 6,558 | -62% | 1 | 1 | 0% | 3,175 | 972 | -69% | 0 | 0 | — |
case-04 | pass→pass | 15,139 | 19,676 | +30% | 1 | 1 | 0% | 2,548 | 2,429 | -5% | 0 | 0 | — |
case-05 | pass→fail | 18,410 | 10,142 | -45% | 1 | 1 | 0% | 3,071 | 2,191 | -29% | 0 | 0 | — |
case-06 | pass→pass | 15,250 | 16,946 | +11% | 1 | 1 | 0% | 2,869 | 3,803 | +33% | 0 | 0 | — |
case-07 | fail→pass | 15,671 | 5,746 | -63% | 1 | 1 | 0% | 2,579 | 1,534 | -41% | 0 | 0 | — |
case-08 | fail→fail | 16,664 | 4,468 | -73% | 1 | 1 | 0% | 3,178 | 1,157 | -64% | 0 | 0 | — |
case-09 | fail→pass | 7,656 | 4,202 | -45% | 1 | 1 | 0% | 1,154 | 1,044 | -10% | 0 | 0 | — |
case-10 | fail→pass | 5,413 | 2,862 | -47% | 1 | 1 | 0% | 791 | 907 | +15% | 0 | 0 | — |
case-11 | fail→pass | 16,809 | 2,147 | -87% | 1 | 1 | 0% | 900 | 848 | -6% | 0 | 0 | — |
case-12 | fail→pass | 12,450 | 6,409 | -49% | 1 | 1 | 0% | 2,130 | 1,642 | -23% | 0 | 0 | — |
case-13 | pass→pass | 20,452 | 12,646 | -38% | 1 | 1 | 0% | 3,336 | 2,154 | -35% | 0 | 0 | — |
case-14 | pass→pass | 20,290 | 18,363 | -9% | 1 | 1 | 0% | 3,246 | 3,746 | +15% | 0 | 0 | — |
case-15 | fail→fail | 6,705 | 5,679 | -15% | 1 | 1 | 0% | 1,184 | 1,444 | +22% | 0 | 0 | — |
case-16 | pass→pass | 19,987 | 21,391 | +7% | 1 | 1 | 0% | 3,203 | 3,735 | +17% | 0 | 0 | — |
case-17 | fail→pass | 15,719 | 8,503 | -46% | 1 | 1 | 0% | 2,510 | 1,866 | -26% | 0 | 0 | — |
case-18 | pass→pass | 14,130 | 8,176 | -42% | 1 | 1 | 0% | 2,288 | 1,976 | -14% | 0 | 0 | — |
case-19 | pass→pass | 16,341 | 18,091 | +11% | 1 | 1 | 0% | 2,432 | 2,872 | +18% | 0 | 0 | — |
case-20 | pass→fail | 15,743 | 7,260 | -54% | 1 | 1 | 0% | 2,777 | 939 | -66% | 0 | 0 | — |
case-21 | pass→pass | 17,274 | 22,784 | +32% | 1 | 1 | 0% | 2,705 | 3,663 | +35% | 0 | 0 | — |
case-22 | pass→pass | 19,275 | 15,203 | -21% | 1 | 1 | 0% | 2,832 | 3,108 | +10% | 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, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 17 comparable cases. 4 cases got worse with the skill loaded, and they are 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.