A terminal-driven browser automation skill that allows AI agents to navigate the web, interact with elements, and manage persistent sessions via Playwright CLI.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clawbrowser
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install clawbrowser using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Clawbrowser is a comprehensive skill designed to empower AI agents with the ability to drive web browsers through the Microsoft Playwright CLI. It enables precise navigation, form interactions, and data extraction without requiring a full graphical interface. By integrating this into the Openclaw Skills ecosystem, developers can automate complex web workflows including session persistence, visual debugging, and automated reporting.
The skill focuses on efficiency by training agents to use element snapshots and session-based configurations. This ensures that browser interactions are reliable, repeatable, and easily auditable. Whether you are building a research agent or a testing bot, Clawbrowser provides the necessary hooks to treat the web as a programmable interface.
To get started with this skill, install the Playwright CLI globally and verify the installation:
npm install -g @playwright/cli@latest
playwright-cli --help
You can configure default behaviors by creating or editing a playwright-cli.json file. Use the playwright-cli install command if browser binaries are missing in your environment.
Clawbrowser maintains a structured approach to browser data and session state. The following table describes how information is organized within the Openclaw Skills framework:
| Component | Format | Description |
|---|---|---|
| Sessions | Directory | Stores cookies, local storage, and indexedDB data for persistent auth. |
| Snapshots | Text/JSON | Maps UI elements to short-hand references (e.g., e1, e2) for the agent. |
| Media | PNG/WebM | Stores screenshots and video recordings of the browser viewport. |
| Traces | Zip/JSON | Contains detailed execution logs and network activity for debugging. |
| Config | JSON | Defines browser type, viewport size, and timeout settings. |
Loading
An automated security checker designed to perform comprehensive audits on code repositories.

A comprehensive EVM wallet management skill for executing crypto transactions and network operations using the cast CLI tool.

A demonstration skill providing a sandbox environment to test and visualize security audit badges within AI agent workflows.

A decentralized research marketplace where AI agents earn coins by fulfilling data bounties and spend them to access high-quality research packages.

A comprehensive toolset for managing EVM wallets and executing blockchain transactions directly through an AI agent via Foundry cast.

A specialized demonstration skill designed to showcase and verify the audit badge functionality within the environment.








































