agentleFS
ncklrsncklrs/claude-chrome-user-testing/skills/critical-paths/SKILL.md

critical-paths

Skill20 starsChanged 10 months agoMITView on GitHub

Test must-work user journeys defined in project configuration. Critical paths are essential flows that must always function correctly. A critical path is a user journey that: - Must work for the site/app to be usable - Represents core business functionality - Would cause significant harm if broken - Should be tested before every deployment

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Test must-work user journeys defined in project configuration.

    line 3
  2. Distinctive:

    Must work for the site/app to be usable

    line 8
  3. Distinctive:

    Should be tested before every deployment

    line 11
  4. Distinctive:

    Verify something exists or is true.

    line 72
  5. Distinctive:

    Wait for a condition.

    line 89
  6. Very specific:

    When --fail-fast is enabled:

    line 240
  7. Distinctive:

    Stop testing after first failure

    line 241
  8. Distinctive:

    Match personas: Use personas that would actually do this journey

    line 331
  9. Distinctive:

    Update regularly: Keep paths in sync with features

    line 332
  10. Distinctive:

    Test the tests: Run paths manually to verify they make sense

    line 333

≈ “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

Critical Paths Skill

Test must-work user journeys defined in project configuration. Critical paths are essential flows that must always function correctly.

What is a Critical Path?

A critical path is a user journey that:

  • Must work for the site/app to be usable
  • Represents core business functionality
  • Would cause significant harm if broken
  • Should be tested before every deployment

Examples of Critical Paths

Site TypeCritical Paths
E-commercePurchase flow, Account creation, Cart
SaaSSignup, Login, Core feature usage
ContentNavigation, Search, Article viewing
BankingLogin, Transfer, Balance check

Path Definition Schema

Full Schema

{
  "paths": [
    {
      "name": "string (required)",
      "description": "string (required)",
      "persona": "persona-id (required)",
      "priority": "critical|high|medium|low (optional, default: high)",
      "timeout": "number in seconds (optional, default: 120)",
      "steps": [
        {
          "action": "navigate|task|verify|wait|input",
          "...action-specific fields"
        }
      ],
      "onFailure": "continue|stop (optional, default: continue)"
    }
  ]
}

Step Actions

navigate

Navigate to a URL path.

{
  "action": "navigate",
  "to": "/path",
  "waitFor": "load|networkidle (optional)"
}

task

Perform a task using the persona's behavior.

{
  "action": "task",
  "do": "description of task to perform",
  "timeout": 30
}

verify

Verify something exists or is true.

// Verify element exists
{ "action": "verify", "element": "description of element" }

// Verify text appears
{ "action": "verify", "text": "exact text to find" }

// Verify URL matches
{ "action": "verify", "url": "/expected-path" }

// Verify URL contains
{ "action": "verify", "urlContains": "confirmation" }

wait

Wait for a condition.

// Wait for seconds
{ "action": "wait", "seconds": 2 }

// Wait for element
{ "action": "wait", "for": "element description" }

// Wait for navigation
{ "action": "wait", "for": "page load" }

input

Enter specific data into a field.

{
  "action": "input",
  "field": "field description or selector",
  "value": "value to enter"
}

Execution Logic

Path Execution

function executePath(path, baseUrl):
  results = { steps: [], status: 'pending' }

  loadPersona(path.persona)
  startTime = now()

  for step in path.steps:
    stepResult = executeStep(step, baseUrl)
    results.steps.append(stepResult)

    if stepResult.status == 'fail':
      results.status = 'fail'
      results.failedAt = step
      captureScreenshot()
      if path.onFailure == 'stop':
        break
      else:
        return results

  results.status = 'pass'
  results.duration = now() - startTime
  return results

Step Execution

function executeStep(step, baseUrl):
  switch step.action:
    case 'navigate':
      navigate(baseUrl + step.to)
      return { status: 'pass' }

    case 'task':
      success = performTaskAsPersona(step.do)
      return { status: success ? 'pass' : 'fail' }

    case 'verify':
      found = verifyCondition(step)
      return { status: found ? 'pass' : 'fail' }

    case 'wait':
      waitForCondition(step)
      return { status: 'pass' }

    case 'input':
      fillField(step.field, step.value)
      return { status: 'pass' }

Priority Handling

Execution Order

  1. Critical paths first
  2. High priority second
  3. Medium priority third
  4. Low priority last

Failure Impact

PriorityOn FailureReport Level
criticalMajor alertCRITICAL
highAlertERROR
mediumWarningWARNING
lowNoteINFO

Report Generation

Data Structure

const pathReport = {
  summary: {
    total: 3,
    passed: 2,
    failed: 1,
    skipped: 0
  },
  results: [
    {
      name: 'purchase-flow',
      status: 'pass',
      duration: 45,
      steps: [
        { action: 'navigate', status: 'pass', duration: 2 },
        { action: 'task', status: 'pass', duration: 15 },
        // ...
      ]
    },
    {
      name: 'password-reset',
      status: 'fail',
      duration: 15,
      failedAt: { step: 4, action: 'task', error: 'Button not clickable' },
      screenshot: 'path-password-reset-fail.png'
    }
  ],
  actionRequired: [
    {
      priority: 'high',
      path: 'password-reset',
      issue: 'Form submission broken',
      recommendation: 'Check submit button handler'
    }
  ]
};

Report Sections

  1. Summary: Pass/fail counts
  2. Results: Each path's outcome
  3. Failures: Detailed failure information
  4. Action Required: Prioritized fix list
  5. Screenshots: Evidence of failures

Integration with CI/CD

Fail-Fast Mode

When --fail-fast is enabled:

  • Stop testing after first failure
  • Report what was tested so far
  • Mark remaining paths as skipped
  • Useful for quick validation

Exit Status Interpretation

All pass + no critical failures = SUCCESS
Any critical failure = FAILURE
Only non-critical failures = WARNING (could be configurable)

Template Path Definitions

E-commerce

{
  "paths": [
    {
      "name": "guest-purchase",
      "description": "Guest checkout flow",
      "persona": "impulse-buyer",
      "priority": "critical",
      "steps": [
        { "action": "navigate", "to": "/" },
        { "action": "task", "do": "find a product" },
        { "action": "task", "do": "add to cart" },
        { "action": "task", "do": "go to cart" },
        { "action": "task", "do": "proceed to checkout" },
        { "action": "task", "do": "fill shipping info" },
        { "action": "verify", "element": "payment form" }
      ]
    },
    {
      "name": "product-search",
      "description": "Search for products",
      "persona": "comparison-shopper",
      "priority": "high",
      "steps": [
        { "action": "navigate", "to": "/" },
        { "action": "task", "do": "search for a product" },
        { "action": "verify", "element": "search results" },
        { "action": "task", "do": "click a result" },
        { "action": "verify", "element": "product details" }
      ]
    }
  ]
}

SaaS Application

{
  "paths": [
    {
      "name": "signup-to-dashboard",
      "description": "New user onboarding",
      "persona": "genz-digital-native",
      "priority": "critical",
      "steps": [
        { "action": "navigate", "to": "/signup" },
        { "action": "task", "do": "complete registration" },
        { "action": "verify", "urlContains": "dashboard" },
        { "action": "verify", "element": "welcome message" }
      ]
    },
    {
      "name": "login-flow",
      "description": "Existing user login",
      "persona": "busy-executive",
      "priority": "critical",
      "steps": [
        { "action": "navigate", "to": "/login" },
        { "action": "input", "field": "email", "value": "test@example.com" },
        { "action": "input", "field": "password", "value": "testpass123" },
        { "action": "task", "do": "submit login" },
        { "action": "verify", "element": "dashboard or main app" }
      ]
    }
  ]
}

Best Practices

  1. Start small: Define 2-3 critical paths first
  2. Be specific: Clear step descriptions help the agent
  3. Match personas: Use personas that would actually do this journey
  4. Update regularly: Keep paths in sync with features
  5. Test the tests: Run paths manually to verify they make sense
  6. Prioritize correctly: Reserve "critical" for truly essential flows

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.