From tac
Runs full project test suite—lint, type check, unit tests, build—stopping on failure. Outputs JSON results sorted by failures for pre-commit validation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/tac:testThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run the project's test suite and report results in structured JSON format.
Run the project's test suite and report results in structured JSON format.
Proactively identify issues in the application before they impact users. This command executes the validation stack and reports results for automated processing.
Execute each validation command in order:
IMPORTANT: If a test fails, stop processing and return results thus far.
Return ONLY a JSON array with test results:
[
{
"test_name": "lint_check",
"passed": true,
"execution_command": "npm run lint",
"test_purpose": "Validates code style and syntax",
"error": null
},
{
"test_name": "type_check",
"passed": true,
"execution_command": "npx tsc --noEmit",
"test_purpose": "Validates TypeScript types",
"error": null
},
{
"test_name": "unit_tests",
"passed": false,
"execution_command": "npm test",
"test_purpose": "Validates core functionality",
"error": "FAIL tests/auth.test.ts - Expected 200, received 401"
}
]
Each test result includes:
| Field | Description |
|---|---|
test_name | Identifier for the test category |
passed | Boolean - true if test passed |
execution_command | Exact command to reproduce |
test_purpose | What this test validates |
error | Error message if failed, null if passed |
Sort the JSON array with failed tests (passed: false) at the top.
This command is the REQUEST phase of a closed loop:
/test → [JSON results] → /resolve-failed-test {result}
The structured output enables automated resolution.
npx claudepluginhub melodic-software/claude-code-plugins --plugin tacSets up validation commands for testing, linting, type-checking, and building in Node.js, Python, Go, Rust, Java projects. Detects stack and generates CI/CD or agent workflow commands.
Binary pass/fail quick gate that runs typecheck, lint, and test commands. Use as a final sanity check after tasks, not for deep analysis.
Verifies code after changes with typecheck, lint, tests, build for Node/TS, Python, Go, Rust projects. Auto-fixes errors, detects secrets, circular deps, deadcode, AI slop.