TestKit plugin
A plugin of testing skills: from requirements and test cases to API, mobile and web runs backed by traceable evidence.
- 101 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Write-method tests and issue submission run only with your permission.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/testkitDetects 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.
- Write-method tests and issue submission run only with your permission.
This is third-party code. Review the repository files before installing.
What it does
TestKit packages repeatable testing workflows as Agent Skills for Claude Code, Codex and compatible agents. The chain runs from requirements analysis, strategy and test points to cases, review, execution, defect retesting and acceptance. API testing has two paths: Arazzo for deterministic business flows and Schemathesis for examples, coverage and fuzzing. For Android, iOS and Web the build, environment and scope are frozen before a run, and results are linked to raw evidence. Further skills analyze logs, review SQL, turn video into an issue and assess performance.
Who it is for. QA engineers and developers who want to run testing through an agent with verifiable artifacts.
Good fit when
- You need test points and cases from a requirement, exported to Excel.
- You import OpenAPI, Swagger, YApi or Postman and automate API checks.
- A device run must keep evidence, not just a pass mark.
- A defect is retested after a fix and acceptance is recorded.
- You need to analyze logs or review SQL for safety.
Not a fit when
- You need one small test without process or artifacts.
- You are not permitted to write to the system under test, inspect code or submit issues.
- The plugin cannot be installed whole under project rules.
Example request
Use TestKit to analyze this coupon requirement, draft test points and cases, ask about missing rules and do not execute tests.Limitations
TestSpec, PerfSpec and execution skills depend on shared directories, so install the plugin as a whole; only five self-contained skills, such as `api-test-automation`, install individually. Excel export needs Python dependencies. The agent asks before each step that changes a system or publishes a result. The README video shows a concept, not a live run. One author maintains the project, so check support for your platform before adopting it in a team process.
How to disable. In Claude Code remove the plugin through `/plugin` and reload plugins; in Codex remove it through the plugin manager. Clean the `testspec/` results directory separately.
FAQ
Can I install a single skill?
Yes, but only the self-contained ones, such as `api-test-automation`: `gh skill install winhok/testkit api-test-automation`. The rest need the plugin's shared directories.
Which clients are supported?
Claude Code and Codex through the plugin; other Agent Skills clients through individual skills and `npx skills`.
Where do results go?
Requirements, analysis, test points, JSON cases and review are saved under `testspec/changes/<name>/`.
What about credentials?
They come from environment variables, and persisted results are redacted.
How do the API paths differ?
Arazzo defines a step scenario, while Schemathesis generates checks from a schema.
Discussion
No questions or reviews yet.
Related
PluginsQA agents, skills and an MCP server: test generation, coverage gaps, flaky tests, security scans and chaos testing
Thirteen testing skills from requirements to report: strategy, cases, API checks, Playwright and flaky-test triage.
MCP server, agent and skills for end-to-end UI tests where the agent drives the computer through vision on cloud machines
A CLI from TestMu AI that writes end-to-end tests from requirements and runs them in a browser or on mobile, with a skill for your agent
A library of 169 QA skills per language: test design, API, UI, regression, load, security and LLM testing.
A skill and tester agent for Claude Code that test the 1C interface through the 1c-testpilot MCP server with evidence
README in short
The README walks from installation to a first coupon task and states the boundaries: code inspection, write-method tests and issue submission need permission.