ccc-test-coverage

/ccc-test-coverage — close the coverage gaps

CC Commander · /ccc-test-coverage · runs the real coverage tool, writes tests for the gaps worst-first, and reports before/after numbers.

Arguments

$ARGUMENTS optionally narrows the scope to a path or glob (for example src/services). Empty means the whole project.

What happens

1. Detect the framework and coverage command. Use the first match:

   Indicator	Coverage command
   vitest.config.*	npx vitest run --coverage
   jest.config.* or a jest key in package.json	npx jest --coverage --coverageReporters=json-summary
   pytest.ini / pyproject.toml with pytest	pytest --cov=src --cov-report=json
   Cargo.toml	cargo llvm-cov --json
   pom.xml with JaCoCo	mvn test jacoco:report
   go.mod	go test -coverprofile=coverage.out ./... then go tool cover -func=coverage.out

   Read the threshold from project config (coverageThreshold, --cov-fail-under, JaCoCo rules, CI config). Use 80% only when the project sets none. If no coverage tool is configured, ask before installing one (picker below).

2. Record the baseline. Run the coverage command, parse the JSON summary or terminal output, and list files below the threshold, worst first. For each, identify:

   • untested functions or methods;
   • missed branches (if/else, switch cases, ternaries, error and catch paths);
   • dead code that inflates the denominator (report it; do not delete it without asking).

3. Write the missing tests, file by file, in this order:

   1. happy path with valid input;
   2. error handling: invalid input, missing data, failed network or I/O;
   3. edge cases: empty arrays, null or undefined, boundary values such as 0, -1 and the maximum;
   4. each remaining branch.

   Rules for every test:

   • place it where the project keeps tests (foo.ts → foo.test.ts by default);
   • copy the project's import style, assertion library and mocking approach;
   • mock only external dependencies (database, APIs, file system, clock);
   • keep tests independent, with no shared mutable state;
   • name tests by behavior, for example create_user_with_duplicate_email_returns_409.