agentleFS
Sign inStart free
pranataa-labspranataa-labs/Workflow-Master/skills/verify-and-test/SKILL.md

verify-and-test

Skill2 starsChanged 8 days agoMITView on GitHub

Evidence-based verification procedure for Stage 6 (Verify). Detects the project stack (Node/TypeScript, Python, Laravel/PHP, Go), runs the available typecheck, lint, test, and build commands, and produces a truthful pass/fail report with real command output. Also defines the AAA test-writing standard and regression-test rules for bug fixes. Use after any code change, before claiming a task is done, before committing, and whenever the user asks to test or verify.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Test terisolasi: database di-reset per test, layanan eksternal (payment, WhatsApp, email, LLM) di-mock.

    line 72

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Verify and Test

Tahap 6 (Verify) mewajibkan bukti eksekusi nyata. Agen dilarang menyatakan "sudah berhasil", "test lulus", atau "sudah diperbaiki" tanpa menjalankan perintah verifikasi dan menampilkan hasilnya.

Rule terkait: 04 (Testing & QA), 06 (Code Quality), 18 (Operational Discipline).

1. Aturan Inti

  1. Tanpa output, tanpa klaim. Setiap klaim keberhasilan wajib disertai perintah yang dijalankan dan ringkasan output-nya.
  2. Jalankan yang tersedia, jangan mengarang. Bila proyek tidak memiliki test suite, laporkan "tidak ada test suite" dan jalankan verifikasi terkuat yang ada (typecheck, build, smoke run).
  3. Gagal berarti berhenti. Bila verifikasi gagal, perbaiki atau laporkan. Jangan melemahkan test, menambah skip, atau menghapus assertion agar lulus.
  4. Bug fix wajib disertai regression test yang gagal sebelum perbaikan dan lulus setelahnya, bila test suite tersedia.
  5. Warning bukan error, tetapi warning baru yang muncul akibat perubahan wajib disebutkan.

2. Verifikasi Otomatis

# Deteksi stack lalu jalankan semua pemeriksaan yang tersedia
powershell -ExecutionPolicy Bypass -File skills/verify-and-test/scripts/verify.ps1

# Hanya tampilkan perintah yang akan dijalankan
powershell -ExecutionPolicy Bypass -File skills/verify-and-test/scripts/verify.ps1 -DryRun

# Proyek di subfolder
powershell -ExecutionPolicy Bypass -File skills/verify-and-test/scripts/verify.ps1 -Path ./frontend

Deteksi yang dilakukan:

PenandaPemeriksaan (bila tersedia)
package.jsonscript typecheck, lint, test, build melalui npm/pnpm/yarn sesuai lockfile; tsc --noEmit bila ada tsconfig.json tanpa script typecheck
pyproject.toml / requirements.txtruff check ., mypy . (bila terkonfigurasi), pytest -q
composer.json + artisanphp artisan test; selain itu vendor/bin/phpunit
go.modgo vet ./..., go test ./...

Skrip keluar dengan kode 1 bila ada pemeriksaan yang gagal.

3. Standar Penulisan Test (AAA)

it("menolak checkout bila keranjang kosong", async () => {
  // Arrange
  const user = await createUser();
  // Act
  const res = await api.as(user).post("/checkout");
  // Assert
  expect(res.status).toBe(422);
  expect(res.body.errors.cart).toBeDefined();
});
  • Satu perilaku per test; nama test menjelaskan perilaku, bukan nama fungsi.
  • Assertion spesifik (toBe(422)), bukan sekadar toBeTruthy().
  • Test terisolasi: database di-reset per test, layanan eksternal (payment, WhatsApp, email, LLM) di-mock.
  • Tanpa sleep tetap; gunakan fake timer atau polling dengan batas waktu.

Prioritas Cakupan

  1. Otorisasi dan batas akses (tamu vs member vs admin).
  2. Perhitungan bisnis (harga, diskon, skor, ranking).
  3. Validasi input dan jalur error.
  4. Happy path alur utama.

4. Verifikasi Visual dan Runtime

Untuk perubahan UI, build saja tidak cukup:

  • Jalankan dev server dan buka halaman yang berubah.
  • Periksa console browser bebas error.
  • Periksa tampilan mobile (lebar 375 px) dan desktop.
  • Untuk notebook atau skrip data, jalankan ujung ke ujung dan periksa output akhir.

5. Format Laporan

[VERIFY] <proyek>
Stack     : node (npm), typescript
typecheck : PASS  (tsc -b, 0 error)
lint      : PASS  (eslint, 2 warning lama)
test      : PASS  (vitest, 48 passed, 0 failed)
build     : PASS  (vite build, 5.2 s)
Manual    : halaman /about dicek di 375 px dan 1440 px, console bersih
Verdict   : TERVERIFIKASI

Bila ada yang gagal, tuliskan FAIL, kutip baris error yang relevan, dan jelaskan langkah berikutnya.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free