Test Automation Skills and Agents plugin
Seven QA agents and ten skills for test automation: Playwright in TypeScript, Selenium in Java, API, accessibility and ISTQB.
- 262 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Agents run browser tests and API calls against the project and edit test files.
- Tests tagged `@destructive` are marked as able to change the state of the environment.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/test-automation-skills-and-agentsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Agents run browser tests and API calls against the project and edit test files.
- Tests tagged `@destructive` are marked as able to change the state of the environment.
This is third-party code. Review the repository files before installing.
What it does
The plugin assembles a QA engineer's workspace: seven agents and ten skills. The agents split roles. `playwright-test-planner` plans scenarios, `playwright-test-generator` writes tests, `playwright-test-healer` repairs failing ones, `selenium-test-specialist` and `api-tester-specialist` cover Selenium and API work, `test-refactor-specialist` tidies up, and `qa-orchestrator` distributes work among them. The skills cover e2e and regression on Playwright, WCAG 2.2 AA accessibility checks, driving a live browser through `playwright-cli`, Selenium in Java with Allure and AssertJ, API testing and ISTQB manual testing. `qa-investigation` hunts the cause of one failing test and keeps an investigation journal in markdown, and `grill-me-qa` interviews the author of an automation plan before implementation. Scoped rules live in `instructions/`: locator priority, no hard waits and exactly one tag per test from `@smoke`, `@sanity`, `@regression`, `@e2e`, `@api`, `@destructive`.
Who it is for. QA engineers and SDETs who write Playwright or Selenium tests and want shared roles and rules for the agent.
Good fit when
- A new feature needs a set of e2e scenarios, with one agent planning and another writing tests.
- Playwright tests fail without an obvious cause and `qa-investigation` works out whether the test is flaky or the app has a real bug.
- The team checks pages against WCAG 2.2 AA with automated tests.
- Manual testing needs a test plan and test cases in the ISTQB manner.
- A Java project needs Selenium rules with explicit waits and AssertJ applied consistently.
Not a fit when
- Tests are written on another stack: the reference patterns cover only Playwright in TypeScript and Selenium in Java.
- You need a single Playwright skill: narrower sets exist as separate repositories.
- Team policy requires manual review of every test edit, while the agents here change files themselves.
Example request
Plan e2e scenarios for the checkout page, generate Playwright tests and repair the ones that fail.Limitations
The repository consists of documentation and templates: it has no build or tests, and its only check is the structural linter `node scripts/lint-skills.mjs`. For GitHub Copilot the files are copied into `.github/agents`, `.github/instructions` and `.github/skills`; for other tools they move into that tool's rules system under the same names. Tests tagged `@destructive` run separately, not in parallel, so they need their own environment. Agents launch a browser and call the project's API, so prepare a test environment in advance and keep secrets out of the repository.
How to disable. Run `/plugin uninstall test-automation-skills-agents@fugazi-test-automation` and drop the marketplace with `/plugin marketplace remove`. Delete copied files from `.github/` by hand.
FAQ
How do I install the plugin in Claude Code?
Run `/plugin marketplace add fugazi/test-automation-skills-agents`, then `/plugin install test-automation-skills-agents@fugazi-test-automation`. All seven agents and ten skills become available.
Can I install a single skill?
Yes, through skills.sh: `npx skills add https://github.com/fugazi/test-automation-skills-agents --skill playwright-e2e-testing`. Swap in any skill name from the README list.
Does the set work in Cursor or Copilot?
Yes, it is described as tool-agnostic. For Copilot the files are copied into `.github/agents` and `.github/skills`; Cursor, Windsurf and others have separate guides in the `docs` directory.
What does the qa-orchestrator agent do?
It distributes work among the specialist agents: planning, generating, healing and refactoring tests. It only reads and searches itself, and the agents it calls make the edits. The others can also be called directly.
What does the one-tag-per-test rule mean?
Each test gets exactly one tag from `@smoke`, `@sanity`, `@regression`, `@e2e`, `@api`, `@destructive`. The tag goes on the test itself, not on `describe()` or the class, and tags are never combined.
Discussion
No questions or reviews yet.
Related
PluginsGuides for agents writing and fixing Playwright tests, with advice on locators, waiting and diagnosing CI failures
A Playwright and TypeScript project template with rules and 17 skills so Claude Code, Codex and Cursor write tests to a standard
57 reference files on Playwright tests in TypeScript: the agent pulls the one that fits the task, from locators to CI sharding
QA agents, skills and an MCP server: test generation, coverage gaps, flaky tests, security scans and chaos testing
A skill that lets an agent write and run Playwright scripts: page tests, forms, screenshots and link checks
A self-improving QA agent: natural-language tests for web and mobile that self-heal and remember runs
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Test Automation Skills and Agents (this card) | no | 262 | |
| Playwright Best Practices Skill | no | 417 | 57 reference files on Playwright only; this one adds Selenium, API, manual testing and seven role agents |
| TestDino Playwright Skill | no | 412 | guides for writing and repairing Playwright tests; this one has a wider scope and an orchestrator that distributes work |
README in short
The author calls the repository a catalog to copy into real projects. Agents set role and boundaries, instructions set shared rules, and skills bring step-by-step workflows with reference files and templates. CI validates file structure on every change.