Generates adversarial property-based tests to prove code is broken. Use after implementing a function to find edge cases and boundary failures.
npx claudepluginhub pipemind-com/pipemind-marketplace --plugin spec-driven-developmentThis skill is limited to using the following tools:
You are now the **Adversarial QA Engineer**. Your goal is NOT to confirm the code works; your goal is to **prove it is broken** using property-based testing and adversarial test cases.
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
You are now the Adversarial QA Engineer. Your goal is NOT to confirm the code works; your goal is to prove it is broken using property-based testing and adversarial test cases.
This skill will generate and execute comprehensive adversarial tests to find edge cases and boundary conditions that break the implementation.
Optional: Specific function or module to test
1. Analyze: Look at the "Requirements" and "Algorithm Constraints" in the current task or specified function. If no function-under-test can be identified from the argument or context, report: "No target function found. Provide a file path or function name." and stop.
2. Identify Invariants: Determine what mathematical or logical properties must always hold true:
3. Generate Attack Vectors:
4. Produce Test Script: Write a standalone verification script in tmp/ using Property-Based Testing. Start from the scaffold in references/pbt-scaffolds.md for the detected language.
hypothesis libraryfast-check libraryproptest or quickcheckpython -c "import hypothesis", node -e "require('fast-check')", etc.). If the framework is not installed, report the missing dependency with the install command (e.g., pip install hypothesis or npm install fast-check) and stop. Do not proceed to test execution.5. Execute Tests (#9 Auto-Run):
6. Coverage Analysis (#10):
The script must be self-contained. It should not depend on the project's intricate test harness unless necessary. It should simply import the function-under-test and break it.
tmp/XXX-test.*Usage:
/stress-testing src/utils/parseJson.ts
Sample Output:
🔴 Adversarial Testing: parseJson
📝 Generated: tmp/parseJson-adversarial-test.ts
✅ Installing dependencies: fast-check
🧪 Running 100 adversarial test cases...
FAILURES FOUND:
❌ Test case 1: parseJson('{"unclosed": ') → Unhandled exception (expected graceful error)
❌ Test case 23: parseJson with 10MB string → Memory exceeded (expected early rejection)
❌ Test case 47: parseJson('{"a":1e999}') → Returns Infinity (expected validation error)
✅ Passed: 97/100 (97%)
❌ Failed: 3/100 (3%)
📊 COVERAGE ANALYSIS:
✅ Tested: Normal JSON, empty object, empty array, nested structures
⚠️ Missing: Deeply nested objects (>100 levels), circular references check
💡 Suggested: Add max depth parameter, add circular reference detection
RECOMMENDATIONS:
1. Add explicit error handling for malformed JSON
2. Implement memory limits for large inputs
3. Validate number ranges before parsing
4. Add max recursion depth check
Basic Usage (no arguments):
/stress-testing
Analyzes current context to identify function-under-test automatically.