agentleFS
agenticlukeagenticluke/browser-qa-plus/skill/SKILL.md

browser-qa

Skill0 starsChanged 2 months agoMITView on GitHub

Use browser tools after a deploy to test page views, user actions, speed, and access needs.

24 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use opensource-pipeline, security-bounty-hunter and 5 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • TironFoxSkill

    opensource-pipeline

    Open-source pipeline: fork, sanitize, and package private projects for safe public release. Chains 3 agents (forker, sanitizer, packager). Triggers: '/opensource', 'open source this', 'make this public', 'prepare for open source'.

    Also in 3 projects that use this

  • RojiosSkill

    security-bounty-hunter

    Hunt for exploitable, bounty-worthy security issues in repositories. Focuses on remotely reachable vulnerabilities that qualify for real reports instead of noisy local-only findings.

    Also in 3 projects that use this

  • Exploration-labsSkill

    token-budget-advisor

    Proactive token budget assessment and task chunking strategy. Use this skill when queries involve multiple large file uploads, requests for comprehensive multi-document analysis, complex multi-step workflows with heavy research (10+ tool calls), phrases like "complete analysis", "full audit", "thorough review", "deep dive", or tasks combining extensive research with large output artifacts. This skill helps assess token consumption risk early and recommend chunking strategies before beginning work.

    Also in 3 projects that use this

  • ElasticSkill

    accessibility

    Accessibility guidance for Kibana. Use this skill when working with or reviewing EUI components, resolving a11y-related (@elastic/eui) ESLint issues, and ensuring proper use of ARIA attributes, focus management, keyboard interactions, and accessible naming conventions.

    Also in 3 projects that use this

  • nilssonrSkill

    seo

    Full SEO workflow for auditing pages and creating SEO-optimized content. Provides checklists with educational explanations aligned with Google Search Central documentation.

    Also in 3 projects that use this

  • Microsoft AzureSkill

    git-workflow

    Squad branching model: dev-first workflow with insiders preview channel

    Also in 3 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

39 rules. 0 are common, 36 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 23 edited versions we read keep these.

  • 36

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 13
  2. Distinctive:

    Ask for these items if they are not known:

    line 24
  3. Distinctive:

    Test login details, if needed.

    line 28
  4. Distinctive:

    Do not use real payment details.

    line 32
  5. Distinctive:

    If a page needs a login and no test login is ready, test all public pages.

    line 34
  6. Distinctive:

    Use an approved browser tool, such as Playwright, Puppeteer, or Claude in Chrome.

    line 36
  7. Distinctive:

    Wait until the main page content is ready.

    line 41
  8. Distinctive:

    Check for browser errors.

    line 42
  9. Very specific:

    Check page calls for 4xx and 5xx errors.

    line 44
  10. Distinctive:

    Check these speed goals when the browser tool can read them:

    line 50
  11. Distinctive:

    If a speed value cannot be read, mark it as not tested.

    line 54
  12. Distinctive:

    Run each speed test more than once when time allows.

    line 56
  13. Distinctive:

    Test the main tasks first.

    line 60
  14. Distinctive:

    Check that each link goes to the right page.

    line 63
  15. Distinctive:

    Test forms with safe and valid data.

    line 64
  16. Distinctive:

    Check the success message or next page.

    line 65
  17. Distinctive:

    Test empty, bad, and very long form values.

    line 66
  18. Distinctive:

    Check that each error is clear and shown near the right field.

    line 67
  19. Distinctive:

    Test login, a private page, and logout when a test login is ready.

    line 68
  20. Distinctive:

    Test key tasks, such as search, sign-up, setup, cart, and checkout.

    line 69
  21. Distinctive:

    Test the Back button after key steps.

    line 70
  22. Distinctive:

    Test loading, empty, error, and success states when they can be reached.

    line 71
  23. Distinctive:

    Stop before any step that could charge money, publish data, or harm live data.

    line 72
  24. Distinctive:

    Do not call a change a bug if it was planned.

    line 103
  25. Distinctive:

    Test dark mode if the site has it.

    line 105
  26. Distinctive:

    Run axe-core or a similar tool on each key page when it is ready.

    line 109
  27. Distinctive:

    Use only the keyboard to:

    line 124
  28. Distinctive:

    Make sure focus does not get stuck.

    line 132
  29. Distinctive:

    Never mark a skipped test as passed.

    line 150
  30. Distinctive:

    Check browser errors and failed page calls.

    line 216
  31. Distinctive:

    Test all main menu links.

    line 218
  32. Distinctive:

    Test sign-up with safe test data.

    line 219
  33. Distinctive:

    Test login with the test login.

    line 220
  34. Distinctive:

    Search for a known item and a missing item.

    line 221
  35. Distinctive:

    Test one full task with only the keyboard.

    line 223
  36. Distinctive:

    Write the report with proof for each failure.

    line 224
  37. Very specific:

    Use /benchmark for a deeper speed test.

    line 228
  38. Very specific:

    Use /canary-watch for checks after a deploy.

    line 229
  39. Distinctive:

    Add this skill to pull request checks for changes to the user interface.

    line 230

“Common” is measured: kept by at least 6 in 10 of the 23 edited versions of this file we could read. “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

Browser QA

Original skill by ECC. Used with clear credit under its open-source license.

When to Use

Use this skill when:

  • A feature is on a test or preview site.
  • You need to check a full page.
  • You want to test before a release.
  • You review a change to the user interface.
  • You need to test phone and tablet views.
  • You need to check access for people with disabilities.

Before You Start

Ask for these items if they are not known:

  • The page URL.
  • The pages and user tasks to test.
  • Test login details, if needed.
  • The allowed test data.
  • Saved images to compare, if any.

Do not use real payment details. Do not send real orders, emails, or messages unless the user says it is safe. Do not delete or change live data.

If a page needs a login and no test login is ready, test all public pages. Mark the login tests as blocked.

Use an approved browser tool, such as Playwright, Puppeteer, or Claude in Chrome.

Phase 1: Basic Checks

  1. Open the target URL.
  2. Wait until the main page content is ready.
  3. Check for browser errors.
  4. Ignore known noise from ads, stats tools, and other outside tools. List what you ignored.
  5. Check page calls for 4xx and 5xx errors.
  6. Ignore calls that were stopped on purpose. Note them in the report.
  7. Take a top-of-page image at these widths:
    • Phone: 375 px
    • Tablet: 768 px
    • Desktop: 1440 px
  8. Check these speed goals when the browser tool can read them:
    • LCP under 2.5 seconds
    • CLS under 0.1
    • INP under 200 ms
  9. If a speed value cannot be read, mark it as not tested. Do not guess.

Run each speed test more than once when time allows. Slow networks and cold starts can change the result.

Phase 2: User Actions

Test the main tasks first.

  1. Open each main menu link.
  2. Check that each link goes to the right page.
  3. Test forms with safe and valid data.
  4. Check the success message or next page.
  5. Test empty, bad, and very long form values.
  6. Check that each error is clear and shown near the right field.
  7. Test login, a private page, and logout when a test login is ready.
  8. Test key tasks, such as search, sign-up, setup, cart, and checkout.
  9. Test the Back button after key steps.
  10. Test loading, empty, error, and success states when they can be reached.
  11. Stop before any step that could charge money, publish data, or harm live data.

For a link that opens a new tab, check the new tab. Then return to the first tab.

Phase 3: Page View Checks

Take full-page images of key pages at:

  • 375 px
  • 768 px
  • 1440 px

For each size, check for:

  • Text cut off or hidden.
  • Images outside the page.
  • Sideways page scroll.
  • Items that cover each other.
  • Missing buttons, text, or images.
  • Menus that cannot open or close.
  • Pop-ups that cannot be closed.
  • Fixed bars that hide page content.
  • Text or controls that are too small to use.

Compare with saved images when they exist. Flag:

  • A move of more than 5 px.
  • A missing item.
  • A new item that should not be there.
  • A clear change in size, space, color, or font.

Do not call a change a bug if it was planned. If the goal is not known, mark it for review.

Test dark mode if the site has it. Test zoom at 200 percent on at least one key page.

Phase 4: Access Checks

Run axe-core or a similar tool on each key page when it is ready.

Check for:

  • WCAG A and AA issues.
  • Low color contrast.
  • Images with no useful text name.
  • Form fields with no label.
  • Buttons or links with no clear name.
  • More than one item with the same ID.
  • A bad heading order.
  • Missing page areas, such as main or nav.
  • Focus that is hard to see.
  • Focus that moves in the wrong order.

Use only the keyboard to:

  1. Move through all controls with Tab and Shift+Tab.
  2. open links and buttons with Enter or Space.
  3. Open and close menus and pop-ups.
  4. Close pop-ups with Escape.
  5. Finish one key task.

Make sure focus does not get stuck. After a pop-up closes, focus should go back to the control that opened it.

A tool scan does not find every access issue. Report both tool results and hand checks.

Failed or Blocked Tests

Retry a failed step once if the cause may be a slow page or tool fault.

For each failed or blocked test, save:

  • The page URL.
  • The page size.
  • The exact test step.
  • What should have happened.
  • What happened instead.
  • An image, browser error, or failed page call.
  • Whether the problem happens again.

Never mark a skipped test as passed.

Output Format

## QA Report: [URL]
Tested: [date and time]
Build: [build name or commit, if known]
Browser: [name and version]
Result: [PASS, FAIL, or BLOCKED]

### Summary
- Passed: [count]
- Failed: [count]
- Blocked: [count]
- Not tested: [count]

### Basic Checks
- PASS: The page loads.
- FAIL: `/api/profile` returns 500.
- NOT TESTED: INP could not be read.
- Images: [file names or links]

### User Actions
- PASS: Main menu links open the right pages.
- PASS: The form shows errors for bad data.
- FAIL: The phone menu does not open.

### Page View Checks
- PASS: The desktop page has no overlap.
- FAIL: The main image goes past the phone screen.
- REVIEW: The title moved 8 px from the saved image.

### Access Checks
- FAIL: 1 WCAG AA color contrast issue.
- PASS: 0 WCAG A issues.
- FAIL: Keyboard focus gets stuck in the sign-up pop-up.

### Bugs
#### BUG-1: Phone menu does not open
- URL: [page URL]
- Page size: 375 x 812
- Steps:
  1. Open the page.
  2. Select the menu button.
- Expected: The menu opens.
- Actual: Nothing happens.
- Proof: [image, error, or failed page call]

### Blocked Tests
- Login test: No test login was ready.

Example Use

Request:

Test https://preview.example.com on phone, tablet, and desktop.
Check sign-up, login, search, and the main menu.
Do not send email or make a real payment.

Run:

  1. Open the home page at 375 px, 768 px, and 1440 px.
  2. Check browser errors and failed page calls.
  3. Save top and full-page images.
  4. Test all main menu links.
  5. Test sign-up with safe test data. Stop before email is sent.
  6. Test login with the test login.
  7. Search for a known item and a missing item.
  8. Run access checks.
  9. Test one full task with only the keyboard.
  10. Write the report with proof for each failure.

Use With Other Skills

  • Use /benchmark for a deeper speed test.
  • Use /canary-watch for checks after a deploy.
  • Add this skill to pull request checks for changes to the user interface.

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.