A bridge that enables Claude Code to interact with the Chrome browser for advanced web automation and data retrieval.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install claude-chrome
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install claude-chrome using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Claude Chrome?
Claude Chrome is a specialized integration designed to extend the capabilities of Claude Code by connecting it directly to a live Google Chrome instance. By utilizing a dedicated browser extension acting as an MCP server, this skill allows the agent to navigate websites, click elements, fill out forms, and extract content in real-time. It provides a powerful alternative to standard web fetching tools within the Openclaw Skills ecosystem, specifically optimized for tasks that require a full browser environment.
This skill is ideal for developers and researchers who need their AI agents to perform complex, multi-step web workflows. Whether it is verifying a UI change, gathering technical documentation from multiple sources, or automating repetitive web-based administrative tasks, Claude Chrome provides the high-fidelity interaction required for modern web applications.
Claude Chrome Use Cases
- Automating multi-step navigation across complex web applications to extract data.
- Verifying frontend UI elements and page content during the development process.
- Performing deep research by following links and summarizing content from various domains.
- Integrating live web data directly into coding workflows within Openclaw Skills.
How Claude Chrome Works
- The skill initiates a command to the Claude Code CLI on the target node using a login shell.
- It establishes a connection to the Chrome browser via the Claude Chrome native host bridge.
- The AI agent processes the natural language prompt and translates it into browser actions like navigation or clicking.
- Using the --chrome and --print flags, the skill executes the task non-interactively to ensure automation compatibility.
- Once the browser task is finished, the results are returned as a system message to the Openclaw Skills interface.
Claude Chrome Setup
To get started, ensure Claude Code is installed and the Claude Chrome extension is active in your browser. Use the following command to check if the bridge is ready:
nodes.run node=<your-node-id> command='["bash", "-lc", "pgrep -f \"claude --chrome-native-host\""]'
Execute a browsing task by running:
nodes.run node=<your-node-id> commandTimeoutMs=120000 command='["bash", "-lc", "claude --dangerously-skip-permissions --chrome -p \"Go to example.com and read the headline\""]'
Claude Chrome Data Schema & Taxonomy
The skill manages interaction data and execution logs to provide transparency into the automation process. The following table describes the data points handled:
| Component | Description | Format |
|---|---|---|
| Command Timeout | Duration allowed for the task to complete | Integer (ms) |
| Native Host PID | The process ID of the Chrome bridge | Number |
| Execution Flags | Parameters like --dangerously-skip-permissions and --print | String |
| Result Output | The final text or summary retrieved from the browser | String |
All data is processed through the Openclaw Skills framework, ensuring that results are delivered even if the initial gateway connection times out.
Claude Chrome Advanced Features
- Support for --dangerously-skip-permissions to enable fully autonomous web interaction without manual approval.
- High-latency task handling where commands continue running in the background until completion.
- Seamless transition between local coding tasks and live web browsing in a single session.
- Specialized benchmarking for different task complexities to optimize timeout settings within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Pndr
Pndr is a comprehensive personal productivity command center that integrates with AI agents via MCP to manage tasks, habits, and journals through natural language.
Agent Communication Skill
A high-performance WebSocket-based communication framework designed for real-time bidirectional messaging and state synchronization between multiple AI agents.
Finance UX Observer
A privacy-centric research tool that silently monitors session transcripts to extract financial behavioral signals and generate redacted UX insight reports.
OpenSea API
A comprehensive integration for querying NFT data, trading on the Seaport marketplace, and performing cross-chain ERC20 token swaps.
Memory Scan
A dedicated security scanner for AI agent memory files and configurations designed to detect prompt injections, credential leaks, and malicious instructions.
Skill-Scan: Security Auditor for Agent Skills
A multi-layered security scanner designed to audit Openclaw Skills for malicious code, prompt injection, and behavioral threats before they are enabled.
Featured*
