chrome-automation — Claude Code skill: what it does, how to install
Controls an active Chrome window via the agent-browser CLI. Useful for extracting web data, filling forms, and replaying DevTools recordings while keeping active login sessions intact.
npx skills add zc277584121/marketing-skills --skill chrome-automation -a claude-codeWhat chrome-automation does
This skill hooks into your main running browser process instead of launching a blank automated window. It inspects existing tabs before navigating anywhere. That approach preserves cookies, authentication states, and open dashboards without forcing you to log in again.
Actions rely on interactive snapshots. The CLI assigns references like @e1 or @e2 to clickable targets and input fields. The agent then clicks elements, inputs text, evaluates JavaScript, or captures screenshots. It can also parse and execute step-by-step recordings exported from Chrome DevTools Recorder in JSON or Puppeteer format.
Chrome must run with remote debugging enabled. Without it, the CLI cannot locate the browser process. On Chrome 144 and newer, remote debugging can switch to WebSocket mode, requiring a direct connection string like ws://127.0.0.1:9222/devtools/browser alongside Node 24 or newer.
What it helps with
- Interact with web applications while preserving existing user logins
- Extract text and table counts directly from rendered pages
- Replay browser action recordings exported from Chrome DevTools
- Capture targeted screenshots to inspect page layout changes
Example prompts for the agent
- Find the open tab with the analytics dashboard and take a screenshot.
- Fill out the contact form on the active page and submit it.
- Replay the steps from recording.json in the current Chrome tab.
How to install chrome-automation
1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill chrome-automation -a claude-code2Manually: put the files in place
~/.claude/skills/chrome-automation/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/chrome-automation/
1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill chrome-automation -a codex2Manually: put the files in place
.agents/skills/chrome-automation/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill chrome-automation -a cursor2Manually: put the files in place
.cursor/skills/chrome-automation/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install zc277584121/marketing-skills chrome-automation2Manually: put the files in place
.github/skills/chrome-automation/SKILL.md1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill chrome-automation -a gemini-cli2Platform CLI
gemini skills install https://github.com/zc277584121/marketing-skills --path skills/chrome-automationManually: put the files in place
.devin/skills/chrome-automation/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill chrome-automation -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about chrome-automation
How to install chrome-automation in Claude Code?
Run npx skills add zc277584121/marketing-skills --skill chrome-automation -a claude-code. The configuration saves to ~/.claude/skills/chrome-automation/SKILL.md.
What does chrome-automation do?
It drives your actual Chrome session via agent-browser. The agent lists active tabs, clicks indexed elements, inputs text, and extracts structured page data.
What prerequisites are needed for chrome-automation?
You need the agent-browser CLI installed and Google Chrome running with remote debugging enabled on port 9222.
