agentleFS
dvy1987dvy1987/agent-loom/.agents/skills/browser-testing-with-devtools/SKILL.md

browser-testing-with-devtools

Skill4 starsChanged 3 months agoMITView on GitHub

Test and debug browser UIs using Chrome DevTools MCP — DOM inspection, console errors, network requests, performance traces, and visual verification. Load when building or debugging anything that renders in a browser, verifying a UI fix, profiling Core Web Vitals in a real page, or the user asks for browser testing with DevTools. Requires chrome-devtools MCP configured. Not for backend-only or CLI work. Pairs with frontend-design and performance-optimization.

62 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use deprecation-and-migration, observability-and-instrumentation and 7 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.
  • HK-hubSkill

    deprecation-and-migration

    Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when deciding whether to maintain or sunset existing code.

    Also in 34 projects that use this

  • HK-hubSkill

    observability-and-instrumentation

    Instruments code so production behavior is visible and diagnosable. Use when adding logging, metrics, tracing, or alerting. Use when shipping any feature that runs in production and you need evidence it works. Use when production issues are reported but you can't tell what happened from the available data.

    Also in 28 projects that use this

  • MrCipherSmithSkill

    constraint-driven-development

    Use when a project needs a persistent, measurable quality bar — starting a project without documented standards, 'set up constraints', or when coverage/perf/a11y thresholds get re-debated per-PR. Produces CONSTRAINTS.md with enforceable numbers, ratchets, and a guard against silent weakening. NOT for throwaway prototypes or a one-off CI setup.

    Also in 12 projects that use this

  • seikaikyoSkill

    doubt-driven-development

    對非平凡決策啟動新鮮 context 對抗審查(找碴不背書),在修正還便宜的時候抓出錯誤方向。適用:高風險改動(production、資安敏感邏輯、不可逆操作)、不熟的程式碼、要宣稱「這樣是安全的 / 可行的」之前。機械性操作與一行修改不適用。

    Also in 28 projects that use this

  • seikaikyoSkill

    interview-me

    需求不明時的意圖萃取訪談:一次一題、每題附上自己的猜測、聽出「真正想要 vs 覺得應該要」,直到能預測使用者反應(約 95% 信心)才動工。適用:需求缺少對象 / 動機 / 成功標準 / 約束,或使用者點名「訪談我」「先確認一下」「我們確定嗎」。明確自足的指示、純資訊查詢、機械性操作不適用。

    Also in 26 projects that use this

  • VercelSkill

    web-design-guidelines

    Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".

    Also in 5 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 56 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use Chrome DevTools MCP to verify runtime behavior — what the user sees, not what static analysis guesses.

    line 23
  2. Distinctive:

    Verify in browser: before claiming a UI fix works.

    line 27
  3. Very specific:

    Use isolated profile by default (--isolated).

    line 28
  4. Very specific:

    Use --autoConnect only when logged-in state is required — document the security trade-off.

    line 48
  5. Distinctive:

    No changes — citation audit passed.

    line 150

“Common” is measured: kept by at least 6 in 10 of the 56 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 Testing with DevTools

Use Chrome DevTools MCP to verify runtime behavior — what the user sees, not what static analysis guesses.

Hard Rules

  • Verify in browser before claiming a UI fix works.
  • Use isolated profile by default (--isolated); never use personal browser for untrusted pages.
  • Read-only JS in page context unless user explicitly approves mutations.
  • Capture evidence: screenshot, console excerpt, network failure, or trace — not vibes.
  • Reproduce from a clean navigation when debugging flaky UI.

Setup (chrome-devtools MCP)

{
  "mcpServers": {
    "chrome-devtools": {
      "command": "npx",
      "args": ["-y", "chrome-devtools-mcp@latest", "--isolated"]
    }
  }
}

Use --autoConnect only when logged-in state is required — document the security trade-off.


Workflow

Step 1 — Define what to verify

State: URL, user action, expected DOM/console/network outcome.

Step 2 — Navigate and observe

  • Screenshot for visual state
  • Console for errors/warnings
  • Network for failed requests or wrong payloads
  • DOM/a11y tree for structure and labels

Step 3 — Diagnose

Map symptom → tool: layout → computed styles; slow paint → performance trace; API bug → network panel.

Step 4 — Fix in code, re-verify

Same steps as Step 2 after change; keep before/after screenshots when useful.

Step 5 — Report

Include evidence snippets and whether issue is fully resolved or partially mitigated.


When NOT to use

  • Pure server/API changes with no browser surface
  • Environments where MCP cannot run (document manual repro instead)

Gotchas

  • Stale service worker caches mask fixes — hard refresh or disable SW when debugging.
  • Flaky tests from animation timing — wait for stable selectors.
  • --autoConnect exposes personal cookies and history.
  • Executing arbitrary JS in page can mutate production-like data.

Common Rationalizations

ExcuseReality
"The code looks correct"Runtime DOM, CSS cascade, and race conditions disagree.
"Unit tests cover it"JSDOM does not run full layout, fonts, or real network.
"I'll check the browser later"Later never comes; verify before marking done.
"MCP is too heavy"One screenshot + console read often saves hours of guesswork.
"It's just a CSS tweak"Tweaks cause CLS and overflow bugs visible only live.

Output Format

## Browser verification — [feature]

URL: [path]
Steps: [actions]
Evidence: [screenshot/console/network]
Result: [pass/fail + notes]

Examples

<examples> <example> <input>"Button click does nothing on /checkout."</input> <output>Console: `TypeError` on click handler; network shows no POST; fix null guard; re-verify POST 200 and success UI.</output> </example> </examples>

Verification

  • Reproduced issue in browser with evidence captured
  • Post-fix verification on same steps
  • No new console errors on happy path
  • Isolated profile used (or autoConnect risk documented)
  • Network/DOM findings tied to specific code change

Red Flags

  • UI fix claimed done without browser verification
  • Personal browser profile used on untrusted pages
  • Page-context JS mutates state without explicit approval
  • Stale service worker cache masks whether fix works

Prune Log

Last pruned: 2026-07-04

  • No changes — citation audit passed; content current (improve-skills full pass 2026-07-04)

Impact Report

URL: [path] | Issue: [one line]
Evidence: [screenshot/console/network]
Status: [fixed/partial/blocked]

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.