メニュー
Design golden path tests that validate critical user journeys end-to-end. Outputs journey mapping, test scope decisions, data strategy, and reliability patterns.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/kalilurrahman/kr-claudiator-skills --skill golden-path-testingコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SOC 職業分類に基づく
SKILL.md を表示中
| name | golden-path-testing |
| description | Design golden path tests that validate critical user journeys end-to-end. Outputs journey mapping, test scope decisions, data strategy, and reliability patterns. |
| argument-hint | ["critical user journeys","tech stack","test framework","environment access"] |
| allowed-tools | Read, Write |
Golden path tests validate the most critical user journeys — the paths that must work for the business to function. Unlike comprehensive E2E tests, golden path tests are intentionally narrow: they cover the core happy paths that, if broken, cause immediate revenue or user impact. They run frequently and must be highly reliable.
## Selection Criteria
A journey qualifies as a golden path if:
1. It generates significant revenue or drives core business value
2. Breaking it would cause immediate user-visible impact
3. It is frequently used (top N% by volume)
4. Its failure would require P1 incident response
## Example Golden Paths by Product Type
E-commerce:
✓ Search → Product Page → Add to Cart → Checkout → Order Confirmation
✓ Login → View Order History → Track Order
✗ Browse related products (nice to have, not revenue-critical)
SaaS:
✓ Sign Up → First Workspace → Create First Project → Invite Teammate
✓ Login → Create Report → Export → Share
✗ Customise profile photo (not revenue-critical)
Banking:
✓ Login (with MFA) → Check Balance → Transfer Funds
✓ Login → Pay Bill → Confirm Payment
✗ Download 2-year statement (infrequent; not critical)
import pytest
from playwright.sync_api import Page, expect
class TestCheckoutGoldenPath:
"""The single most critical user journey — must always pass."""
@pytest.fixture(autouse=True)
def setup(self, page: Page, base_url: str):
self.page = page
self.base_url = base_url
def test_complete_purchase_flow(self):
"""
Golden Path: Guest checkout — product to order confirmation.
If this fails, revenue generation is impaired.
SLA: Must pass 99.9% of runs (max 1 failure per 1000).
"""
page = self.page
# Step 1: Search for product
page.goto(f"{self.base_url}/")
page.fill('[data-testid="search-input"]', "blue widget")
page.press('[data-testid="search-input"]', "Enter")
expect(page.locator('[data-testid="product-card"]').first).to_be_visible(timeout=5000)
# Step 2: View product
page.locator('[data-testid="product-card"]').first.click()
expect(page.locator('[data-testid="product-title"]')).to_be_visible()
expect(page.locator('[data-testid="price"]')).to_contain_text("$")
# Step 3: Add to cart
page.click('[data-testid="add-to-cart"]')
expect(page.locator('[data-testid="cart-count"]')).to_have_text("1")
# Step 4: Checkout
page.click('[data-testid="checkout-btn"]')
page.fill('[data-testid="email"]', "test@goldenpath.invalid")
page.fill('[data-testid="card-number"]', "4242424242424242")
page.fill('[data-testid="card-expiry"]', "12/28")
page.fill('[data-testid="card-cvc"]', "123")
page.click('[data-testid="place-order"]')
# Step 5: Confirm success
expect(page.locator('[data-testid="order-confirmed"]')).to_be_visible(timeout=15000)
order_id = page.locator('[data-testid="order-id"]').text_content()
assert order_id, "Order ID not displayed"
# Golden path tests must be held to higher reliability standards
# than regular E2E tests — they must almost never fail due to test issues
# Reliability patterns:
class ReliableStep:
MAX_RETRIES = 3
RETRY_DELAY_MS = 500
@staticmethod
def click_with_retry(page, selector: str):
"""Retry click for transient DOM timing issues."""
for attempt in range(ReliableStep.MAX_RETRIES):
try:
page.locator(selector).click(timeout=5000)
return
except Exception as e:
if attempt == ReliableStep.MAX_RETRIES - 1:
raise
page.wait_for_timeout(ReliableStep.RETRY_DELAY_MS)
@staticmethod
def fill_with_retry(page, selector: str, value: str):
page.locator(selector).fill("") # Clear first
page.locator(selector).fill(value)
# Verify the value was entered correctly
actual = page.locator(selector).input_value()
assert actual == value, f"Failed to fill {selector}: expected {value!r}, got {actual!r}"
## Test Data Approaches
APPROACH 1: Dedicated Test Accounts
Pros: Stable; no interference with real users
Cons: Test data accumulates; accounts may expire
Pattern:
- Pre-created accounts: test-golden-path@company.com
- Reset state before each run via API
- Separate test product catalog (real prices, fake inventory)
APPROACH 2: Generated Data per Run
Pros: No state accumulation; fully isolated
Cons: Slower; more setup
Pattern:
- Create test user via API (not UI)
- Use `testuser+<uuid>@company.com` email pattern
- Clean up after test (or let TTL expire)
APPROACH 3: Fixture Data (Recommended for Speed)
Pros: Fast; predictable
Cons: Must maintain fixture state
Pattern:
- Database seed applied before test run
- Test uses known fixture IDs
- Transaction rollback after test
## What NOT to Use
- Production user accounts (GDPR, data contamination)
- Real payment credentials (use test card numbers: 4242 4242 4242 4242)
- Real email addresses (use `@goldenpath.invalid` domain)
# Golden path tests: run frequently, fast feedback
- name: Golden Path Tests
run: pytest tests/golden_path/ -v --timeout=60
env:
BASE_URL: https://staging.example.com
HEADLESS: "true"
# Run on every merge to main AND every 15 minutes in production (synthetic monitoring)
# Alert when golden path fails
- name: Alert on failure
if: failure()
run: |
curl -X POST $SLACK_WEBHOOK_URL -d '{"text": "GOLDEN PATH FAILURE on ${{ github.ref }}: ${{ github.run_url }}"}'
| Anti-Pattern | Problem | Fix |
|---|---|---|
| Too many golden paths | Suite becomes slow; flaky | Max 5-10 golden paths per product area |
| Testing edge cases in golden path | Slow; defeats the purpose | Golden path = happy path only |
| No SLA for golden path reliability | Flakiness tolerated | <0.1% failure rate from test issues required |
| Production data in tests | GDPR; data contamination | Dedicated test accounts or synthetic data |
| Only running on PR | Production can degrade between PRs | Run on schedule (every 15 min) as synthetic monitoring |
The sections below extend this skill into a complete operating playbook so it can run end-to-end inside Claude Code, CoWork, or any agentic tool without further prompting. Pull only the sections you need for a given engagement.
Before producing any output, the skill confirms:
If any of these are missing, the skill asks targeted clarifying questions before generating output. It never invents constraints the user did not state.
The canonical workflow for Golden Path Testing runs in five stages. Each stage has an explicit exit criterion so the skill knows when to advance.
Stage 1 — Frame. Restate the problem in one paragraph. Name the decision, the deadline, the stakeholders, and the success metric. Surface assumptions explicitly so they can be challenged.
Stage 2 — Diagnose. Inventory the current state with concrete evidence: metrics, quotes, screenshots, configs, tickets. Separate facts from interpretations. Identify the two or three root causes that explain most of the gap, not the long tail of symptoms.
Stage 3 — Design. Generate at least two viable options. For each option, capture: what changes, who owns it, what it costs, what it unblocks, what it risks, and how it could fail. Recommend one with a written rationale.
Stage 4 — Execute. Convert the chosen option into a sequenced plan: milestones, owners, dependencies, gating checks, communication cadence, and rollback triggers. Anything that cannot be assigned an owner and a date is not yet a plan.
Stage 5 — Validate. Define how success will be measured, when the measurement happens, and what action follows each possible result. Schedule the retrospective before the work starts, not after.
Depending on the request, the skill returns one or more of:
Every artifact uses clear headings, short paragraphs, and tables where comparison helps. No filler. No restating the prompt. No hedging language when a recommendation is warranted.
The skill applies the following heuristics when choices are not obvious:
| Anti-pattern | Why it fails | What the skill does instead |
|---|---|---|
| Generic best-practice list with no context | Reader cannot act on it | Tailors recommendations to the stated constraints |
| Recommendation without trade-offs | Hides the cost of being wrong | Names the price paid for the recommendation |
| Plan with no owners or dates | Cannot be executed or tracked | Assigns a named owner and a date to every action |
| Metrics theatre | Measures activity, not outcome | Ties every metric back to the user or business outcome |
| Boil-the-ocean scope | Nothing ships | Cuts scope to the smallest valuable slice |
| Buried recommendation | Reader misses the point | Leads with the recommendation in the first paragraph |
The skill self-checks each output against these gates before returning it:
If any gate fails, the skill rewrites the section before returning it.
Context: Design golden path tests that validate critical user journeys end-to-end. Outputs journey mapping, test scope decisions, data strategy, and reliability patterns
Frame: the team needs a defensible recommendation within five working days; the audience is a cross-functional steering group; the cost of delay is higher than the cost of being slightly wrong.
Diagnose: the dominant constraint is decision latency, not analytical depth. Existing data is sufficient for a directional call.
Design: two viable options surfaced. Option A optimises for speed and reversibility. Option B optimises for completeness but slips the deadline by two weeks.
Execute: Option A recommended. Plan sequenced into a two-week sprint with named owners, a mid-point checkpoint, and a clear rollback trigger.
Validate: success measured against a single leading indicator at day 30 and a single lagging indicator at day 90. Retrospective scheduled for day 35.
A one-shot artifact rarely changes outcomes. The skill recommends a lightweight cadence to keep the work alive: