Agent skill
agent-browser
Headless browser automation for AI agents - auto-installs if needed
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/agent-browser-hathbanger-fridas-choppers-gtm
SKILL.md
Agent Browser
Headless browser automation CLI for AI agents. GitHub
On Skill Invoke
Step 1: Check if installed
which agent-browser
Step 2: If NOT installed, offer to install
agent-browser not installed.
This tool lets me browse the web, fill forms, take screenshots, and scrape dynamic content.
Install it?
npm install -g agent-browser
agent-browser install
[Yes] [No]
If user says yes, run:
npm install -g agent-browser && agent-browser install
On Linux, if install fails with dependency errors:
agent-browser install --with-deps
After install, confirm:
Done. agent-browser ready.
Try: agent-browser navigate https://example.com
Step 3: If already installed, show status
agent-browser is installed.
Quick commands:
agent-browser navigate <url> # Go to URL
agent-browser snapshot # Get page structure (for AI)
agent-browser click <ref> # Click element
agent-browser type <ref> <text> # Type into input
agent-browser screenshot # Capture page
agent-browser pdf # Generate PDF
Need help with something specific?
What It Does
Agent-browser gives Claude the ability to:
- Navigate and interact with web pages
- Click, type, scroll
- Take screenshots and generate PDFs
- Extract accessibility trees (semantic page understanding)
- Monitor network requests
- Manage cookies and storage
Common Commands
| Command | What It Does |
|---|---|
agent-browser navigate <url> |
Go to a URL |
agent-browser snapshot |
Get accessibility tree (for AI understanding) |
agent-browser click <ref> |
Click an element by reference |
agent-browser type <ref> <text> |
Type into an input |
agent-browser screenshot |
Capture the page |
agent-browser pdf |
Generate PDF |
When to Use
Use agent-browser when you need to:
- Scrape dynamic content (SPAs, JS-rendered pages)
- Fill out forms programmatically
- Test user flows
- Capture screenshots for docs/previews
- Extract data from authenticated pages
Typical Workflow
agent-browser navigate <url>- load the pageagent-browser snapshot- understand page structureagent-browser click <ref>/agent-browser type <ref> <text>- interactagent-browser screenshotor extract data
Troubleshooting
"chromium not found"
agent-browser install
Permission errors on Linux
agent-browser install --with-deps
Page not loading
Some sites block headless browsers. The tool handles most cases, but very aggressive bot detection may block it.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?