code-generate-tests

Generate tests for untested functions and edge cases, then verify they pass.

Process

1. Identify context:

   • If code changes are involved: run jj diff -s first to see changed files; then use jj diff -- path to restrict to specific files/directories
   • If the user specified specific files or paths, focus on those
   • Understand which functions/modules need test coverage

2. Detect test framework:

   • Invoke code-analyze-project skill to detect test framework and testing patterns
   • Check for existing test files and naming conventions
   • Identify test helper functions and fixtures in use

3. Identify untested code paths:

   • Use Grep to find functions without corresponding tests
   • Check for edge cases, error paths, boundary conditions
   • Identify critical paths that lack coverage

4. Research and analyze:

   • Research testing best practices for the language/framework (use WebSearch/WebFetch)
   • Identify critical paths, edge cases, and error handling scenarios

5. Generate test code:

   • Match existing test conventions (naming, structure, fixtures)
   • Cover happy path, edge cases, and error scenarios
   • Include proper assertions and test organization

6. Run the generated tests:

   • Invoke test skill to run the new tests
   • Analyze any failures

7. Fix test failures:

   • Identify root cause of failures
   • Fix test code or generated code as appropriate
   • Re-run tests to verify

Stop Condition

• If test generation or fixing fails twice, stop and ask for guidance.

Output

1. Test Framework Detected
2. Untested Functions/Paths Identified
3. Generated Tests (with file locations and coverage summary)
4. Test Results (from running tests)
5. Failures Fixed (if any, with explanations)
6. Verification - All new tests pass