Detects test framework and runs the project test suite.
A shared support tool that detects the project's test framework and runs the test suite. Used by agent-hale (self-check), agent-ellis (QA verification), and agent-page (pre-release).
This skill is triggered when the user's prompt contains run tests or test-runner.
Auto-detect the test runner by checking for configuration files:
| Config File | Runner | Command |
|---|---|---|
package.json with test script | npm | npm test |
pytest.ini / pyproject.toml | pytest | pytest |
setup.cfg with [tool:pytest] | pytest | python -m pytest |
go.mod | go test | go test ./... |
Makefile with test target | make | make test |
Cargo.toml | cargo | cargo test |
pom.xml | maven | mvn test |
build.gradle / build.gradle.kts | gradle | gradle test |
Gemfile with rspec / minitest | bundler | bundle exec rspec |
mix.exs | mix | mix test |
If multiple frameworks are detected, run all of them. If none are detected, report that no test framework was found and stop.
Execute the detected test command(s). Capture output including:
Return a structured report:
__TEST_RESULT__
Framework: <name>
Command: <command>
Status: PASS / FAIL
Total: <n>
Passed: <n>
Failed: <n>
Skipped: <n>
Duration: <time>
__TEST_RESULT__
If tests fail, include the failure details (file, test name, error message).
Available to: agent-hale, agent-ellis, agent-page, agent-ross
When invoked by other agents, always emit the __TEST_RESULT__ block.
The calling agent decides how to act on the results.