start-ui
Skill0 starsChanged 6 months agoMITView on GitHub
Pre-flight with Chrome. Runs /start then confirms Chrome DevTools tools are available for UI testing. Use instead of /start when you need browser access.
In this repository14 other files it gives its agents.
- auditSkill
- promptSkill
- protocol-brainsSkill
- protocol-pinkySkill
- protocolSkill
- review-ai-compatSkill
- review-authSkill
- review-cicdSkill
- review-designSkill
- review-monorepoSkill
- review-prSkill
- reviewSkill
- review-test-strategySkill
- startSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 11
Run the standard /start pre-flight, then verify Chrome DevTools tools are loaded and navigate to the frontend.
- Distinctive: line 15
Do not create more nodes to avoid thinking.
- Distinctive: line 17
Create nodes only when they reduce confusion.
- Very specific: line 27
Run ToolSearch to load Chrome DevTools tools:
- Distinctive: line 36
If Chrome tools are not available, tell the user:
- Very specific: line 62
DO NOT read project docs — /start already handles that.
- Very specific: line 63
Keep output under 5 extra lines beyond /start.
≈ “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.
Files
Start UI - Pre-Flight with Chrome
Purpose
Run the standard /start pre-flight, then verify Chrome DevTools tools are loaded and navigate to the frontend.
Principle
Do not create more nodes to avoid thinking.
Create nodes only when they reduce confusion.
Instructions
Step 1 - Run /start
Invoke the start skill to do the normal pre-flight (CodeGraph, branch check, tool loading).
Step 2 - Verify Chrome tools
Run ToolSearch to load Chrome DevTools tools:
ToolSearch: "+chrome-devtools screenshot"
ToolSearch: "+chrome-devtools navigate"Then try mcp__chrome-devtools__list_pages to confirm the connection is live.
If Chrome tools are not available, tell the user:
"Chrome tools not available. Run
claude mcp listto verify thechrome-devtoolsMCP server is registered."
Step 3 - Navigate to the frontend
Once Chrome is confirmed connected, navigate to the frontend app:
mcp__chrome-devtools__navigate_page(url: "http://localhost:3000")Take a screenshot to confirm the app loaded and auth is active. If the page redirects to the login page instead of the app, tell the user:
"Frontend redirected to login. Please authenticate in the browser, then re-run
/start-ui."
Step 4 - Output
Append to the /start output:
**Chrome:** ✅ Connected — navigated to localhost:3000 / ❌ Not availableConstraints
- DO NOT read project docs —
/startalready handles that. - Keep output under 5 extra lines beyond
/start.
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
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.

