Test Runner

Run tests and return structured results.

Input

scope: string  # Optional: directory, file path, or test name pattern

Process

Detect framework from project files (if not provided):
- spec/ + _spec.rb + .rspec = RSpec
- test/ + _test.rb = Minitest
- jest.config.js or *.test.js = Jest/Vitest
- go.mod + _test.go = Go test
- pytest.ini or conftest.py = pytest
Run tests (scoped if provided, otherwise full suite)
Parse output for pass/fail counts and failure details

Framework Commands

Framework	Run All	Run File	Run Single
RSpec	`bundle exec rspec`	`bundle exec rspec path/to/spec.rb`	`bundle exec rspec path:LINE`
Minitest	`bundle exec rake test`	`ruby -Itest path/to/test.rb`	`ruby -Itest path -n test_name`
Jest	`npx jest`	`npx jest path/to/test.js`	`npx jest path -t "name"`
pytest	`pytest`	`pytest path/to/test.py`	`pytest path::test_name`
Go	`go test ./...`	`go test ./path/`	`go test ./path/ -run TestName`

Output

status: PASS | FAIL
total: int
passed: int
failed: int
failures:        # Only if failed
  - file: string
    line: int
    name: string
    message: string

Error Handling

Scenario	Action
Test framework unclear	Ask user
Tests fail	Return FAIL with failure details
Command error	Return FAIL with stderr

test-runner

Test Runner

Input

Process

Framework Commands

Output

Error Handling