메뉴
Frontmatter is fine and the description is well over 40 characters, but the body below the closing `
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Run a project's automated tests end-to-end and return a single structured report (ASCII dashboard + machine-readable trailing lines + path to .test-runs/<UTC-ts>/run.log). Trigger when the user asks to run, execute, or check the status of a project's tests — phrases like "run the unit tests", "run the tests", "execute the test suite", "ai check tests for me", "run npm test and tell me what failed", or "how are the tests doing?". This is the meta-skill that delegates to layer-specific runners (Phase 1 ships only the `unit-test-runner` child); future phases will add `browser-test`, `integration-test-runner`, and `coverage-reporter` to the same orchestration. Do NOT trigger on "tail the log", "show me the latest test output", or "what does .test-runs/latest contain?" — those are plain `Read` of `<project>/.test-runs/latest/run.log`. Do NOT trigger on lint, typecheck, build, deploy, or coverage-only requests — those belong to other (separate) skills. Do NOT trigger on browser-only / Playwright-with-page.goto navi
too short
Folder is `name-mismatch/` but the frontmatter declares `name: totally-different-name`. Linter check 6 must catch this. Long enough to clear the 40-char gate.
Centralize test-run logs under <project>/.test-runs/<UTC-timestamp>/ so every other testing skill (auto-test, unit-test-runner, browser-test, integration-test-runner, flaky-detector, coverage-reporter) writes into a single shared run folder. Use ONLY as a dependency invoked by another testing skill — it is not user-facing on its own. Trigger when another skill needs to (a) start a new run folder via `scripts/init-run.sh`, (b) append per-layer log lines via `scripts/append-log.sh`, or (c) finalize a run via `scripts/finalize-run.sh`. Do NOT trigger on user phrases like "show me the logs", "tail the logs", "check what tests printed" — those should call `auto-test` (which internally uses this skill) or read `<project>/.test-runs/latest/run.log` directly with the Read tool. Do NOT use this skill for non-test logging (application logs, build output, lint output) — those belong to other tools.
Detect the JavaScript / TypeScript unit-test framework in a project (Jest, Vitest, bun:test, Mocha, or @playwright/test used as a unit-test runner) and prepare a canonical run command. Trigger when another testing skill — usually `auto-test` — needs to know "what unit-test framework does this repo use, what command should I run, and which JS runtime / package manager (npm, pnpm, yarn, bun) drives it?". Concretely fires on user phrases like "run the unit tests", "what framework do these tests use?", "detect the test runner for this project", or as a sub-step of `auto-test`. Do NOT trigger on "run end-to-end tests" / "browser tests" / "integration tests" / "Playwright e2e tests with a dev server" — those belong to the Phase 2 `browser-test` skill (this skill's `playwright-runner` framework keyword is reserved for `@playwright/test` used WITHOUT browser navigation, i.e. as a generic test runner). Do NOT trigger on Python (pytest), Rust (cargo test), Go (`go test`), Ruby (rspec), or Elixir tests — those are Phase
Phase-0 spike probe used to verify Claude Code skill discovery and triggering. Trigger ONLY when the user message contains the literal phrase "hello-test ping" — no paraphrases, no partial matches. Output is a single fixed acknowledgement string with the current ISO-8601 UTC timestamp, produced by running a one-liner Bash command. Do NOT trigger on the words "hello", "test", "ping" individually, on phrases like "hello world", "say hello", "ping me", or any other variant. This skill exists purely as a fingerprint for the auto-test-skills repo Phase 0 spike to confirm that user-level skills dropped at ~/.claude/skills/<name>/SKILL.md are auto-discovered. Safe to delete after Phase 0 sign-off.
| name | no-body |
| description | Frontmatter is fine and the description is well over 40 characters, but the body below the closing `---` delimiter is empty — should fail check 8. |
