Test

Run tests with context-efficient output.

Task

$ARGUMENTS

Instructions

Output: ## Test

State: Utility command. Does not modify STATE.md phase.

1. Parse Arguments

Arguments format: [framework] [path] [extra flags]

Framework	Command	Fail-fast
pytest	`pytest -x`	`-x`
jest	`jest --bail`	`--bail`
vitest	`vitest --bail`	`--bail`
go	`go test -failfast`	`-failfast`
npm	`npm test`	(none)
custom	as provided	(none)

Default: pytest if no framework specified.

2. Execute with run_silent

Source and use the run_silent wrapper:

bash

source .claude/skills/implement/scripts/run_silent.sh
run_silent_with_count "tests" "pytest -x tests/" "pytest"

This produces:

•Success: ✓ tests (45 tests, in 2.3s)
•Failure: Full diagnostic output

3. Report Result

On success:

markdown

## Test
✓ [framework] passed ([count] tests)

On failure:

markdown

## Test
✗ [framework] failed

[Error output]

Examples

code

/test                     # pytest -x (default)
/test pytest tests/api    # pytest -x tests/api
/test jest                # jest --bail
/test go ./...            # go test -failfast ./...
/test npm                 # npm test
/test "make test"         # custom command

Constraints

•Always use fail-fast flags to minimize output
•Show full output only on failure
•Don't add verbose flags unless explicitly requested

Exit Criteria

Test result reported with minimal context consumption.