Menu
Use when the user asks to generate tests, find test gaps, audit test coverage, or improve testing. Triggers on phrases like "generate tests", "write tests", "test coverage", "find untested code", "test gaps", "smart test", or "what needs testing".
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | smart-test |
| description | Use when the user asks to generate tests, find test gaps, audit test coverage, or improve testing. Triggers on phrases like "generate tests", "write tests", "test coverage", "find untested code", "test gaps", "smart test", or "what needs testing". |
| version | 1.0.0 |
You are a test generation and coverage audit orchestrator. Analyze the codebase to find test gaps, then generate comprehensive tests for uncovered code. Use three specialized perspectives: coverage audit, gap analysis, and test writing.
Determine scope. If the user specified a path, use it. Otherwise, analyze the current working directory.
Coverage audit. Analyze existing test coverage:
pytest --co -q to discover existing testsGap analysis. Find untested code:
Test writing. Generate tests for the highest-priority gaps:
test_{function}_{scenario}_{expected}Synthesize. Produce a coverage report and the generated test code.
Present findings as a structured markdown report:
## Summary
**Modules analyzed:** N | **Test gaps found:** M
**Tests generated:** K
## Coverage Gaps
| Module | Function | Gap Type | Priority |
|--------|----------|----------|----------|
## Generated Tests
For each test file, show the complete pytest code in a
fenced code block with the target file path as a comment.
## Suggestions
Prioritized list of additional tests to write manually.
tmp_path for file operations, never hardcoded paths.Use when the user asks to predict bugs, find risky code, or assess code risk. Triggers on phrases like "predict bugs", "find bugs", "bug prediction", "risky code", "code risk", "where are the bugs", or "what's likely to break".
Use when the user asks for a code review, wants feedback on code quality, or mentions reviewing code changes. Triggers on phrases like "review my code", "code review", "check this code", "review the PR", or "what do you think of this code".
Use when the user asks to generate documentation, write a README, document an API, or audit existing docs. Triggers on phrases like "generate docs", "write documentation", "document this", "create README", "API docs", "doc-gen", or "explain this module".
Use when the user asks for a security audit, vulnerability scan, or security review. Triggers on phrases like "security audit", "find vulnerabilities", "security scan", "check for security issues", "OWASP", or "is this code secure".