Best Testing Skills for Claude Code
Skills for software testing: test-first development, Playwright end-to-end automation, coverage-driven test generation, browser debugging through Chrome DevTools, root-cause analysis when a run breaks, and chaos engineering for deliberate failure injection. This is the stack that decides whether your agent ever ships code it has not actually run.
Designs chaos experiments, creates failure injection frameworks, and facilitates game day exercis…
Excellent chaos engineering skill: task-oriented workflow, strong safety guards, rollback verification, and well-organized progressive references; only missing When-Not boundary in description.
Use when implementing any feature or bugfix, before writing implementation code
Model discipline skill: Iron Law, full RED-GREEN-REFACTOR cycle with mandatory fail/pass verification, rationalization and red-flag tables, worked bugfix example and completion checklist; supporting doc sits at root rather than references/.
Use when writing E2E tests with Playwright, setting up test infrastructure, or debugging flaky br…
Well-built Playwright E2E skill with correct-vs-brittle selector contrast, full POM example, and a flaky-test debug loop ending in a repeat-each-10 stability check; no When-Not in description.
Complete browser automation with Playwright. Auto-detects dev servers, writes reusable test scrip…
Well-crafted tool wrapper with dev-server detection, reusable script conventions, current Playwright locator patterns and a verify-before-claiming-success rule; layout is flat with API_REFERENCE.md at root instead of references/.
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectu…
Comprehensive QA specialist with ten focused references, failure classification and flakiness handling in the workflow, and strong MUST/MUST-NOT constraints; properly credits adapted TDD content; no explicit When-Not.
Tests in real browsers via Chrome DevTools MCP. Use when building or debugging anything that runs…
Excellent browser-testing skill with three structured debugging workflows, a test-plan template, and an unusually strong security section covering profile isolation, untrusted-content rules, and JS execution constraints.
Guides systematic root-cause debugging. Use when tests fail, builds break, behavior doesn't match…
Excellent systematic debugging skill: six-step ordered triage with decision trees, error-type playbooks, regression-test guard, verification checklist, and a security section treating error output as untrusted data.
Toolkit for interacting with and testing local web applications using Playwright. Supports verify…
Compact official Playwright testing toolkit with a good decision tree, server lifecycle helper and reconnaissance pattern; description is minimal and the workflow is thinner than sibling office skills.
Score comparison
How the top 8 compare across the six evaluation dimensions (0–10). Full written rationale for every score is on each skill's page.
| Skill | Overall | Trigger | Structure | Workflow | Content | Engineering | Security |
|---|---|---|---|---|---|---|---|
| chaos-engineer | 9.2 | 8.0 | 9.3 | 9.2 | 9.0 | 10.0 | 10.0 |
| test-driven-development | 9.1 | 8.5 | 8.0 | 9.2 | 9.0 | 10.0 | 10.0 |
| playwright-expert | 9.0 | 8.0 | 10.0 | 8.0 | 9.0 | 10.0 | 10.0 |
| playwright-skill | 9.0 | 8.5 | 8.0 | 8.4 | 10.0 | 10.0 | 10.0 |
| test-master | 9.0 | 9.0 | 9.3 | 8.0 | 9.0 | 9.0 | 10.0 |
| browser-testing-with-devtools | 8.8 | 9.0 | 7.3 | 8.0 | 9.0 | 10.0 | 10.0 |
| debugging-and-error-recovery | 8.6 | 8.0 | 7.3 | 8.0 | 9.0 | 10.0 | 10.0 |
| webapp-testing | 8.4 | 6.0 | 9.3 | 8.0 | 9.0 | 9.0 | 10.0 |
FAQ
What is a testing skill for Claude Code?
A package that teaches your agent a testing practice rather than a tool: write the failing test before the implementation, drive a real browser to prove a flow works, generate tests from coverage gaps, or trace a failure back to its root cause instead of patching the first plausible symptom.
TDD or Playwright — which do I need?
Usually both. The TDD skill governs the order you write code in and applies at the unit level in any language. The Playwright skills (playwright-skill, playwright-expert) automate a real browser for end-to-end flows — test infrastructure, selectors, CI wiring. TDD keeps the logic honest; Playwright proves the assembled product actually works.
Do these replace my test suite or CI?
No. They run inside your agent session — writing tests, executing them, reading the failures back. Your CI still has the final say, and tests the agent writes should land in the repo like any other code and be reviewed the same way.
What is a chaos skill doing on a testing page?
chaos-engineer (9.2) designs failure-injection experiments — the controlled version of "what happens when this dependency dies." That is testing applied to resilience rather than correctness, and it belongs in the same toolkit.
Every skill here was scored on the same six-dimension rubric — the full scorecard with written rationale is on each skill's page. See more in the Developer Tools skill collection, or browse all evaluated skills.