| name | Claude Code Chrome Browser |
| description | Enable browser automation capabilities in Claude Code using the Chrome extension. This allows agents to interact with web pages, fill forms, click buttons, and perform automated browser tasks. |
| version | 1.0.0 |
| category | automation |
| skillType | mcp |
| triggers | ["browser","chrome","web automation","browse","click","fill form","navigate"] |
| tags | ["browser","automation","chrome","mcp","web"] |
| execution | {"type":"mcp-tool","mcpTool":{"toolName":"mcp__claude-in-chrome__*","defaultParams":null}} |
| runtime | {"runtime":"claude-code","flags":["--chrome"]} |
| notices | [{"type":"requirement","message":"This skill requires the Claude Code Chrome extension to be installed in your browser."},{"type":"info","message":"This skill is only available when using Claude Code as the AI runtime."}] |
Claude Code Chrome Browser
This skill enables browser automation capabilities through the Claude Code Chrome extension.
Capabilities
When this skill is enabled, agents can:
- Navigate to web pages
- Read page content and structure
- Fill out forms and input fields
- Click buttons and links
- Take screenshots
- Execute JavaScript on pages
- Interact with dynamic web applications
Usage
The browser automation tools are available through MCP (Model Context Protocol) when Claude Code is started with the --chrome flag.
Available Tools
mcp__claude-in-chrome__navigate - Navigate to URLs
mcp__claude-in-chrome__read_page - Read page accessibility tree
mcp__claude-in-chrome__find - Find elements using natural language
mcp__claude-in-chrome__form_input - Fill form fields
mcp__claude-in-chrome__computer - Mouse and keyboard interactions
mcp__claude-in-chrome__javascript_tool - Execute JavaScript
mcp__claude-in-chrome__get_page_text - Extract page text content
Requirements
- Install the Claude Code Chrome extension from the Chrome Web Store
- Keep Chrome running with the extension active
- Ensure Claude Code is started with the
--chrome flag (Crewly handles this automatically when this skill is enabled)
Best Practices
- Always wait for page loads before interacting with elements
- Use
find to locate elements by description before clicking
- Take screenshots to verify page state when debugging
- Handle dynamic content with appropriate waits