How do I install the Unit Testing skill?

Run npx skillmds@latest add javierbrea/unit-testing in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Unit Testing safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Unit Testing?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Unit Testing free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Unit Testing?

javierbrea (@javierbrea) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Unit Testing by javierbrea — Agent Skill

Name: Unit Testing
Author: javierbrea

Unit Testing

Complete or maximize unit test coverage for a specific TypeScript file in this repo. Use when: a test run is short of the 100% coverage threshold and you need to find and close the gap, or when deciding whether an uncovered branch is genuinely unreachable. For everyday test-writing conventions (AAA, mocking, placement, type safety), see .agents/rules/unit-testing.md instead — it loads automatically for any *.spec.ts file.

javierbrea 76be62e 2.3 KB Updated Sep 17, 2026

File contents

Unit Testing — closing coverage gaps

All packages enforce 100% coverage (branches, functions, lines, statements) via jest.config.js's coverageThreshold.global; any drop below fails the build. For AAA structure, mocking, placement, and type safety, follow .agents/rules/unit-testing.md — this skill only covers the coverage-completion workflow. For the general Nx-vs-direct command choice, see the root AGENTS.md.

Completing coverage for a specific file

Run the target test file with coverage to get a baseline:
```
cd packages/<package-name>
pnpm test:unit <test-file>
```
Find missing branches by inspecting packages/<package-name>/coverage/lcov.info for BRDA entries with zero hits. Map uncovered branches to exact conditions in the source.
Add tests to hit uncovered branches. Re-run after each change. Repeat until resolved or justified.
Prefer adding tests over suppressing coverage.
If uncovered logic appears obsolete, do not change production code — present evidence and request user confirmation first.

Handling unreachable or defensive branches

When a branch is genuinely unreachable at runtime and cannot be exercised through tests, mark it with an istanbul ignore comment that justifies why:

/* istanbul ignore next -- Defensive guard: invalid parser state is impossible through public API */

Rules for istanbul ignore:

Always include a justification after the -- separator.
Prefer making code non-optional over ignoring it. If a default is never exercised, remove it instead of adding an ignore comment.
Only use for genuinely unreachable code. If a branch can be triggered through a public API, write a test for it.
Keep the ignore scope as narrow as possible. Place on the specific line or branch, not on entire functions.

---name: unit-testingdescription: Complete or maximize unit test coverage for a specific TypeScript file in this repo. Use when: a test run is short of the 100% coverage threshold and you need to find and close the gap, or when deciding whether an uncovered branch is genuinely unreachable. For everyday test-writing conventions (AAA, mocking, placement, type safety), see .agents/rules/unit-testing.md instead — it loads automatically for any *.spec.ts file.---# Unit Testing — closing coverage gapsAll packages enforce **100% coverage** (branches, functions, lines, statements) via `jest.config.js`'s `coverageThreshold.global`; any drop below fails the build. For AAA structure, mocking, placement, and type safety, follow `.agents/rules/unit-testing.md` — this skill only covers the coverage-completion workflow. For the general Nx-vs-direct command choice, see the root `AGENTS.md`.## Completing coverage for a specific file1. **Run the target test file with coverage** to get a baseline: ```bash cd packages/<package-name> pnpm test:unit <test-file> ```2. **Find missing branches** by inspecting `packages/<package-name>/coverage/lcov.info` for `BRDA` entries with zero hits. Map uncovered branches to exact conditions in the source.3. **Add tests** to hit uncovered branches. Re-run after each change. Repeat until resolved or justified.4. **Prefer adding tests over suppressing coverage.**5. If uncovered logic appears obsolete, **do not change production code** — present evidence and request user confirmation first.## Handling unreachable or defensive branchesWhen a branch is genuinely unreachable at runtime and cannot be exercised through tests, mark it with an `istanbul ignore` comment that **justifies why**:```typescript/* istanbul ignore next -- Defensive guard: invalid parser state is impossible through public API */```Rules for `istanbul ignore`:1. **Always include a justification** after the `--` separator.2. **Prefer making code non-optional over ignoring it.** If a default is never exercised, remove it instead of adding an ignore comment.3. **Only use for genuinely unreachable code.** If a branch can be triggered through a public API, write a test for it.4. **Keep the ignore scope as narrow as possible.** Place on the specific line or branch, not on entire functions.

Frequently asked questions