steel-browser Documentation

Open Source Browser API for AI Agents & Apps. Steel Browser is a batteries-included browser sandbox that lets you automate the web without worrying about...

7,000 stars TypeScript TypeScript 1 file · ~5,717 tokens #ai#ai-agents#ai-tools#browser-automation#llm
RAW Doc

Docker & Self-Hosted Deployment Verified OCI

Deploy steel-browser on any server, homelab, or cloud container platform.

1-Click Terminal Command
bash (docker run) 
docker run -d --name steel-browser -p 3000:3000 --restart always steel-dev/steel-browser:latest
Configured Services & Ports
⚙️ api
image: build locally
🔌 Ports: 3000:3000, 9223:9223
⚙️ ui
image: build locally
🔌 Ports: 5173:80
docker-compose.yml
yaml (docker-compose.yml) 
services:
  api:
    build:
      context: .
      dockerfile: ./api/Dockerfile
      args:
        NODE_VERSION: 22.13.0
    ports:
      - "3000:3000"
      - "9223:9223"
    environment:
      - DOMAIN=${DOMAIN:-localhost:3000}
      - CDP_DOMAIN=${CDP_DOMAIN:-localhost:9223}
    volumes:
      - ./.cache:/app/.cache
    networks:
      - steel-network

  ui:
    build:
      context: .
      dockerfile: ./ui/Dockerfile
    ports:
      - "5173:80"
    environment:
      - API_URL=${API_URL:-http://api:3000}
    depends_on:
      - api
    networks:
      - steel-network

networks:
  steel-network:
    name: steel-network
    driver: bridge

1. Project Overview & Quickstart (steel-dev/steel-browser)

File: README.md

**Steel**

Get Started · Documentation · Website · Cookbook

✨ Highlights

Steel.dev is an open-source browser API that makes it easy to build AI apps and agents that interact with the web. Instead of building automation infrastructure from scratch, you can focus on your AI application while Steel handles the complexity.

Under the hood, it manages sessions, pages, and browser processes, allowing you to perform complex browsing tasks programmatically without any of the headaches:

  • Full Browser Control: Uses Puppeteer and CDP for complete control over Chrome instances -- allowing you to connect using Puppeteer, Playwright, or Selenium.
  • Session Management: Maintains browser state, cookies, and local storage across requests
  • Proxy Support: Built-in proxy chain management for IP rotation
  • Extension Support: Load custom Chrome extensions for enhanced functionality
  • Debugging Tools: Built-in request logging and a UI to view/debug sessions with
  • Anti-Detection: Includes stealth plugins and fingerprint management
  • Resource Management: Automatic cleanup and browser lifecycle management
  • Browser Tools: Exposes APIs to quick convert pages to markdown, readability, screenshots, or PDFs.

For detailed API documentation and examples, check out our API reference or explore the Swagger UI directly at http://0.0.0.0:3000/documentation.

Steel is in public beta and evolving every day. Your suggestions, ideas, and reported bugs help us immensely. Do not hesitate to join in the conversation on Discord or raise a GitHub issue. We read everything, respond to most, and love you.

If you love open-source, AI, and dev tools, we're hiring across the stack!

Make sure to give us a star ⭐

🛠️ Getting Started

The easiest way to get started with Steel is by creating a Steel Cloud account. Otherwise, you can deploy this Steel browser instance to a cloud provider or run it locally.

⚡ Quick Deploy

If you're looking to deploy to a cloud provider, we've got you covered.

Deployment methods Link
Pre-built Docker Image (combined API + UI)
1-click deploy to Railway
1-click deploy to Render

💻 Running Locally

Docker

The simplest way to deploy/run a Steel browser instance locally is to run the pre-built Docker image:

bash
# Pull and run the Docker image
docker run -p 3000:3000 -p 9223:9223 ghcr.io/steel-dev/steel-browser

This will start the Steel browser server on port 3000 (http://localhost:3000) and the UI at http://localhost:3000/ui. The 9223 port is used for the console debugger.

You can now create sessions, scrape pages, take screenshots, and more. Jump to the Usage section for some quick examples on how you can do that.

Alternatively, you can run the API and UI separately with docker compose:

bash
docker compose up

For Mac Silicon users, you will need to pass this env flag to the Docker compose command to run the images on the correct platform:

bash
DOCKER_DEFAULT_PLATFORM=linux/arm64 docker compose up

Quickstart for Contributors

When developing locally, you will need to run the docker-compose.dev.yml file instead of the default docker-compose.yml file so that your local changes are reflected. Doing this will build the Docker images from the api and ui directories and run the server and UI on port 3000 and 5173 respectively.

bash
docker compose -f docker-compose.dev.yml up

You will also need to run it with --build to ensure the Docker images are re-built every time you make changes:

bash
docker compose -f docker-compose.dev.yml up --build

If you run on a custom host, create a .env file (see docs/DEVELOPMENT_SETUP.md for variables) or modify the environment variables used by docker-compose.dev.yml to use your host.

Node.js

Alternatively, if you have Node.js and Chrome installed, you can run both the server and the UI directly:

bash
npm install
npm run dev

This will also start the Steel server on port 3000 and the UI on port 5173.

Make sure you have the Chrome executable installed and in one of these paths:

  • Linux:
    /usr/bin/google-chrome

  • MacOS:
    /Applications/Google Chrome.app/Contents/MacOS/Google Chrome

  • Windows:

    • C:\Program Files\Google\Chrome\Application\chrome.exe OR
    • C:\Program Files (x86)\Google\Chrome\Application\chrome.exe

Custom Chrome Executable

If you have a custom Chrome executable or a different path, you can set the CHROME_EXECUTABLE_PATH environment variable to the path of your Chrome executable:

bash
export CHROME_EXECUTABLE_PATH=/path/to/your/chrome
npm run dev

For more details on where this is checked look at api/src/utils/browser.ts.

🏄🏽‍♂️ Usage

If you're looking for quick examples on how to use Steel, check out the Cookbook.

Alternatively you can play with the REPL package too cd repl and npm run start

There are two main ways to interact with the Steel browser API:

  1. Using Sessions
  2. Using the Quick Actions Endpoints

In these examples, we assume your custom Steel API endpoint is http://localhost:3000.

The full REST OpenAPI documentation can be found on our site and on your local Steel instance at http://localhost:3000/documentation.

Using the SDKs

If you prefer to use the our Python and Node SDKs, you can install the steel-sdk package for Node or Python.

These SDKs are built on top of the REST API and provide a more convenient way to interact with the Steel browser API. They are fully typed, and are compatible with both Steel Cloud and self-hosted Steel instances (changeable using the baseURL option on Node and base_url on Python).

For more details on installing and using the SDKs, please see the Node SDK Reference and the Python SDK Reference.

Sessions

The /sessions endpoint lets you relaunch the browser with custom options or extensions (e.g. with a custom proxy) and also reset the browser state. Perfect for complex, stateful workflows that need fine-grained control.

Once you have a session, you can use the session ID or the root URL to interact with the browser. To do this, you will need to use Puppeteer or Playwright. You can find some examples of how to use Puppeteer and Playwright with Steel in the docs below:

Creating a Session using the Node SDK

typescript
import Steel from 'steel-sdk';

const client = new Steel({
  baseURL: "http://localhost:3000", // Custom API Base URL override
});

(async () => {
  try {
    // Create a new browser session with current API fields
    const session = await client.sessions.create({
      blockAds: true,
      proxyUrl: "user:pass@host:port", // optional
      dimensions: { width: 1280, height: 800 }, // optional
    });
    console.log("Created session with ID:", session.id);
  } catch (error) {
    console.error("Error creating session:", error);
  }
})();

Creating a Session using the Python SDK

python
import os
from steel import Steel

client = Steel(
    base_url="http://localhost:3000",  # Custom API Base URL override
)

try:
    # Create a new browser session with custom options
    session = client.sessions.create(
        block_ads=True,
        proxy_url="user:pass@host:port",  # optional
        dimensions={"width": 1280, "height": 800},  # optional
    )
    print("Created session with ID:", session.id)
except Exception as e:
    print("Error creating session:", e)

Creating a Session using Curl

bash
# Launch a new browser session
curl -X POST http://localhost:3000/v1/sessions \
  -H "Content-Type: application/json" \
  -d '{
    "proxyUrl": "user:pass@host:port",
    "blockAds": true,
    "dimensions": { "width": 1280, "height": 800 }
  }'

Selenium Sessions

Note: This integration does not support all the features of the CDP-based browser sessions API.

For teams with existing Selenium workflows, the Steel browser provides a drop-in replacement that adds enhanced features while maintaining compatibility. You can simply use the isSelenium option to create a Selenium session:

typescript
// Using the Node SDK
const session = await client.sessions.create({ isSelenium: true });
python
# Using the Python SDK
session = client.sessions.create(is_selenium=True)

Using Curl

bash
# Launch a Selenium session
curl -X POST http://localhost:3000/v1/sessions \
  -H "Content-Type: application/json" \
  -d '{
    "isSelenium": true
  }'

The Selenium API is fully compatible with Selenium's WebDriver protocol, so you can use any existing Selenium clients to connect to the Steel browser. For more details on using Selenium with Steel, refer to the Selenium Docs.

Quick Actions API

The /scrape, /screenshot, and /pdf endpoints let you quickly extract clean, well-formatted data from any webpage using the running Steel server. Ideal for simple, read-only, on-demand jobs:

Scrape a Web Page

Extract the HTML content of a web page.

bash
# Example using the Actions API
curl -X POST http://0.0.0.0:3000/v1/scrape \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com",
    "delay": 1000
  }'

Take a Screenshot

Take a screenshot of a web page.

bash
# Example using the Actions API
curl -X POST http://0.0.0.0:3000/v1/screenshot \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com",
    "fullPage": true
  }' --output screenshot.png

Download a PDF

Download a PDF of a web page.

bash
# Example using the Actions API
curl -X POST http://0.0.0.0:3000/v1/pdf \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com"
  }' --output output.pdf

Get involved

Steel browser is an open-source project, and we welcome contributions!

  • Questions/ideas/feedback? Come hangout on Discord
  • Found a bug? Open an issue on GitHub

License

Apache 2.0

Made with ❤️ by the Steel team.

File: docs/README.md

Steel Browser Documentation

Welcome to the Steel Browser documentation! This directory contains comprehensive guides and references to help you understand, use, and contribute to Steel Browser.

📚 Documentation Overview

Getting Started

Architecture & Design

API Reference

For New Contributors

  1. Start with the Contributing Guide
  2. Set up your development environment using the Development Setup Guide
  3. Read the Architecture Overview to understand the system
  4. Check out issues labeled good first issue

For Plugin Developers

  1. Read the Plugin Development Guide
  2. Study the Architecture Overview for system understanding
  3. Browse existing plugins in api/src/services/cdp/plugins/
  4. Join our Discord for plugin development discussions

For Users

  1. Check the main README for basic usage
  2. Browse the Steel Cookbook for examples
  3. Use the Troubleshooting Guide if you encounter issues
  4. Visit the API Documentation for detailed API reference

🛠️ Documentation Structure

text
docs/
├── README.md                  # This file - documentation overview
├── ARCHITECTURE.md           # System architecture and design
├── DEVELOPMENT_SETUP.md      # Development environment setup
├── PLUGIN_DEVELOPMENT.md     # Plugin creation guide
└── TROUBLESHOOTING.md        # Common issues and solutions

📖 External Resources

Official Resources

Learning Resources

🤝 Contributing to Documentation

We welcome contributions to improve our documentation! Here's how you can help:

Reporting Documentation Issues

  • Missing Information: If you can't find what you're looking for
  • Outdated Content: If documentation doesn't match current behavior
  • Unclear Instructions: If steps are confusing or incomplete
  • Broken Links: If links don't work or point to wrong resources

Improving Documentation

  1. Fork the repository and create a feature branch
  2. Make your changes to the relevant documentation files
  3. Test your changes by following the instructions you've written
  4. Submit a pull request with a clear description of improvements

Documentation Standards

  • Use clear, concise language that's accessible to all skill levels
  • Include code examples where applicable
  • Add screenshots for UI-related documentation
  • Keep examples up-to-date with current API and features
  • Cross-reference related sections to help users navigate

🔍 Finding What You Need

By Use Case

I want to...

By Component

I'm working with...

📋 Documentation Roadmap

Planned Additions

  • Deployment Guide - Production deployment instructions
  • Security Guide - Security best practices and configuration
  • Performance Guide - Optimization and scaling recommendations
  • Testing Guide - Comprehensive testing strategies
  • Migration Guide - Upgrading between versions
  • FAQ - Frequently asked questions

Community Contributions Needed

  • Platform-specific guides (Windows, macOS, Linux variations)
  • Integration examples with popular tools and frameworks
  • Video tutorials for complex setup procedures
  • Translated documentation for non-English speakers
  • Real-world use case studies

💡 Getting Help

If you can't find what you're looking for in the documentation:

  1. Search existing issues on GitHub
  2. Ask in Discord for real-time help
  3. Create a documentation issue describing what's missing
  4. Check the Steel Cookbook for practical examples

🎯 Documentation Goals

Our documentation aims to be:

  • Comprehensive - Covering all aspects of Steel Browser
  • Accessible - Understandable by users of all skill levels
  • Up-to-date - Reflecting the current state of the project
  • Practical - Including real-world examples and use cases
  • Community-driven - Improved through user feedback and contributions

Happy learning and building with Steel Browser! 🚀

Last updated: [Current Date] - If you notice outdated information, please let us know!

2. Official Technical Reference & Guides (steel-dev/docs)

Steel Docs

Documentation for Steel.dev. Built with Fumadocs and enhanced with custom components.

Contributing

Documentation Structure

Component-Based (/tools/*): Features organized by tool capabilities

Example structure:

text
/tools/steel/
├── Overview/                 # Introduction and setup
├── Steel CLI/             # Command-line features
├── Steel JS SDK/          # Testing and programmatic usage
└── Integrations/             # External connections

Writing Guidelines

After installing dependencies, install Vale, sync the pinned style
packages, and run the prose checks before submitting documentation changes:

bash
brew install vale
bun add --global [email protected]
bun run prose:sync
bun run prose

Vale checks MDX-aware editorial and Steel terminology rules. cspell remains the source of truth
for spelling and project vocabulary. During the initial rollout, Vale reports warnings without
failing the command or CI.

Follow our content rhythm pattern:

text
Paragraph (2-3 sentences) → Code Example → Paragraph → List/Table → Code
  • 60% code, 40% text - Show, don't just tell
  • Never put two paragraphs in a row - Break with code/lists/tables
  • Every concept needs a code example
  • Use complete, contextual sentences - Explain "why" before "how"

Available Components

Layout Components

mdx
<Steps>
  <Step>
    ### Action verb phrase Context sentence explaining what and why. ```terminal
    $ command with options ```
  </Step>
</Steps>

---

<TerminalPicker storage="macOs" flags="-w">

```terminal !! macOS
$ brew install @steel-dev/cli
```

```terminal !! Windows
$ winget install @steel-dev/cli
```

</TerminalPicker>

---

<CodeTabs storage="languageSwitcher">

```typescript
const client = Steel()
const result = await client.sessions.create()
```

```javascript
const client = Steel()
const result = await client.sessions.create()
```

</CodeTabs>

Directives (Auto-converted to components)

mdx
:::callout
type: tip|info|warn|help
### Optional Title
Content here
:::

:::objectives
- What you'll learn
- Another learning objective
:::

:::prerequisites
- Required knowledge
- Tools needed
:::

:::next-steps
- [Next Guide](/path): Description
- [Another Guide](/path): Description
:::

Interactive Components

mdx
<Accordion>
  <AccordionItem>
    <AccordionTrigger>Question</AccordionTrigger>
    <AccordionContent>Answer</AccordionContent>
  </AccordionItem>
</Accordion>

Code Block Features

Languages: terminal, typescript, javascript, json, yaml, console, package-install

Flags: -n (line numbers), -c (copy button), -w (word wrap), -a (animate), -f <file-name> (file name)

Annotations:

mdx
```ts
// !mark[3:5]     Highlight lines 3-5
// !mark[/regex/] Highlight pattern
// !diff +        Addition
// !diff -        Removal
// !collapse[3:5] Collapse lines
```
text
**Special blocks**:

```terminal
$ command         # Renders with prompt
```

```package-install
@steel-dev/cli  # Auto-generates npm/yarn/pnpm commands
```

Page Templates

Overview Page

mdx
---
title: Tool Name
sidebarTitle: Overview
---

## Overview

Expand on what the tool does and its value. Include llms.txt link if available.

## Key features

- **Feature** - User benefit description
- **Feature** - Technical capability explained simply

## Installation

<TerminalPicker>...</TerminalPicker>

:::next-steps

- [Quickstart](/path): Get started in minutes
  :::

use the isLink: true in the front matter to add a link icon to the right of the navbar icon
use the isSeperator: true in the frontmatter to make a doc element look like a seperator (Bold/All caps and spaced out, good for special links)

Quickstart Page

mdx
---
title: Quickstart
---

Brief intro paragraph about what you'll build.

:::objectives

- Learning objective 1
- Learning objective 2
  :::

:::prerequisites

- Prerequisites
  :::

## Quickstart

<Steps>[3-5 focused steps]</Steps>

:::next-steps

- [Next Guide](/path): Description
  :::

File Structure

  • /content/docs/ - All documentation content
  • /components/ - Custom components
  • /public/ - Images and static assets
  • meta.json files control sidebar organization

Quick Checklist

  • Follow content rhythm (no consecutive paragraphs)
  • 60% code, 40% text ratio
  • Every concept has a code example
  • Use appropriate page template
  • Specify language for all code blocks
  • Include context before code
  • Keep paragraphs to 2-3 sentences

Explore Related Technologies & Topics

View All 9,111+ Repositories →

Discover similar high-velocity repositories, agent skills, and OpenAPI specifications across the ecosystem.

#ai Browse ai catalog → #ai-agents Browse ai-agents catalog → #ai-tools Browse ai-tools catalog → #browser-automation Browse browser-automation catalog → #llm Browse llm catalog → ⚡ MCP Servers Model Context Protocol → 🧠 Agent Skills Prompt & rule files → 🔌 OpenAPI Specs REST & Swagger schemas →