Cloud-based AI browser automation for executing complex web tasks through a simple API and natural language instructions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install browser-use-api
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 browser-use-api 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 Browser Use?
The Browser Use skill empowers developers to perform high-level web automation without the overhead of managing local browser drivers. By integrating this capability into Openclaw Skills, users can trigger AI-driven browsing sessions that handle everything from simple data extraction to complex multi-step navigation. It operates via a cloud API, meaning it can bypass environment-specific restrictions and execute tasks in a clean, remote environment.
This skill is designed for scenarios where traditional scraping fails. By utilizing Large Language Models to drive the browser, it can interpret web page structures dynamically, fill out forms, and navigate through interactive elements just as a human would. This makes it an essential component for any automated workflow requiring robust web interaction.
Browser Use Use Cases
- Automating complex multi-step web workflows that require logical decision-making.
- Scraping data from highly dynamic websites or those with advanced anti-bot protections.
- Executing automated form filling and submissions across various web platforms.
- Generating visual audits or verification via screenshots of specific web interaction steps.
- Scaling web tasks when local browser control is unavailable or resource-intensive.
How Browser Use Works
- The user provides a plain English description of the web task to be performed.
- A POST request is sent to the Browser Use API with the task description and API key.
- The cloud infrastructure spins up a browser instance and uses an LLM to navigate the target site.
- The system polls the task status, tracking the progress through states like pending, started, and finished.
- Upon completion, the skill retrieves a structured payload containing the result, action logs, and associated costs.
Browser Use Setup
To get started with this skill within Openclaw Skills, you must first configure your environment with a valid API key.
# Set your API key in the environment
export BROWSER_USE_API_KEY="your_api_key_here"
You can then use the included helper script to run tasks immediately:
# Execute a task and wait for the result
./scripts/browser-use.sh "Go to hacker news and get the top 3 stories"
Browser Use Data Schema & Taxonomy
The skill returns structured JSON data representing the lifecycle and outcome of the browser task. The metadata taxonomy includes:
| Key | Description |
|---|---|
| id | The unique UUID for the task session. |
| status | Current lifecycle state (pending, started, finished, failed). |
| output | The final text result extracted by the AI. |
| steps | A detailed array of actions taken, including screenshots. |
| cost | The exact cost of the transaction in USD. |
| isSuccess | A boolean flag indicating if the goal was achieved. |
Browser Use Advanced Features
- Custom LLM selection, allowing users to specify models like gemini-1.5-flash for specific performance needs.
- Asynchronous execution support with a no-wait flag for background processing in Openclaw Skills.
- Real-time credit monitoring to manage API spend and operational budget.
- Detailed step-by-step action logging for debugging complex AI reasoning during web navigation.
- Stop-task functionality to terminate long-running or runaway browser sessions instantly.
SKILL.md
Loading
Related Openclaw Skills
ToneClone
ToneClone is a specialized AI writing tool that captures and replicates your unique voice for authentic, personalized content generation.
ToneClone CLI
A command-line interface for generating AI content in your authentic voice using trained personas and specific knowledge context.
Praxis Google Workspace CLI
A powerful CLI wrapper for official Google APIs providing direct access to Gmail, Calendar, and Drive without third-party proxies.
FormPass Submit
FormPass Submit provides a structured trust layer for AI agents to discover, validate, and submit data to web forms via an authenticated API.
Database Operations
A professional-grade database specialist for designing schemas, optimizing complex SQL queries, and managing zero-downtime migrations.
.NET Expert
A senior-level AI assistant for building high-performance .NET 9 applications using Clean Architecture and the Result pattern.
Featured*
