n8n Browser-Use Nodes
https://github.com/browser-use/n8n-nodes-browser-use- Category
- AI Agents
- Rank
- No. 1450Tools index
- Pricing
- Open Source
- Type
- TOOL
- Builder
- browser-use
- GitHub
- 45 stars
- Latest release
- v1.2.1
- Date
About
n8n community nodes for the Browser-Use browser-agent framework, letting you orchestrate browser automation flows from n8n.
What it does
It gives n8n a single control surface for Browser Use Cloud. You can submit natural-language tasks, wait for results, request schema-shaped output, reuse browser state, inspect messages, and stop sessions. A separate browser mode provisions managed browsers and returns live-view and CDP connection URLs for other automation tools.
Why it's ranked here
The package covers both established task workflows and newer session-based automation without splitting credentials or nodes. Backward compatibility, structured extraction, reusable sessions, explicit cost limits, and useful error propagation make it practical. Its value is narrower if you do not already use n8n and Browser Use Cloud.
What's good
The API selector preserves older task flows while exposing sessions and standalone browsers. Session reuse keeps cookies and state available for follow-up work. JSON Schema output fits downstream automation better than free-form text. Profiles, persistent workspaces, geographic proxies, recordings, message history, timeouts, and cost caps give operators meaningful control.
Tradeoffs
This is a connector to a managed cloud service, so it requires a Browser Use account, API key, and network access. Waiting workflows poll until completion or terminal failure, which can occupy an n8n execution for long tasks. Direct browser control still belongs to separate tools connected through the returned CDP URL.
How to use it well
Use it when n8n already coordinates research, extraction, form filling, testing, or multi-step web work. Keep existing automations on v2 when they depend on task-specific controls, and choose v3 for reusable sessions or standalone browsers. Pair structured schemas with downstream validation. Choose another tool for local browser execution or direct Playwright-style control.
Technical notes+
The package is TypeScript compiled to CommonJS and ES2019 under strict settings in tsconfig.json. package.json declares the public npm package, Node.js >=18.10, pnpm >=9.1, an n8n-workflow peer dependency, and compiled entries under dist. nodes/BrowserUse/BrowserUse.node.ts combines v2 and v3 properties behind addApiVersionDisplayOptions, while nodes/BrowserUse/BrowserUseV3.ts implements session and browser resources, polling through n8n-workflow sleep, terminal-state handling, NodeApiError wrapping, structured schemas, and session options. credentials/BrowserUseApi.credentials.ts stores the API key as a password field, sends X-Browser-Use-API-Key, and normalizes v3 credential URLs to the compatible v2 endpoint for testing. gulpfile.js copies icons and node metadata. .github/workflows/publish.yml lints, builds, and publishes publicly with npm provenance.
Observed
- License
- MIT
- Primary language
- TypeScript
- Install surface
- n8n community node manager or public npm package n8n-nodes-browser-use-cloud
- Interface
- Single n8n node connecting to Browser Use Cloud API v2 and v3
- Runtime
- Node.js 18.10 or newer; pnpm 9.1 or newer
- Browser interoperability
- Standalone browser creation returns live-view and CDP URLs for external controllers
- Build output
- CommonJS targeting ES2019, with compiled distribution files published
Read from README.md, package.json, index.js, gulpfile.js, credentials/BrowserUseApi.credentials.ts, nodes/BrowserUse/BrowserUseV3.ts, nodes/BrowserUse/BrowserUse.node.ts, LICENSE, .npmignore, CHANGELOG.md, tsconfig.json, .claude/settings.local.json, .github/workflows/publish.yml, nodes/BrowserUse/BrowserUse.node.json.
What it can do
Navigate to web pages
URL or web address → Loaded web page in browser
Fill out web forms automatically
Form data and target form fields → Completed web form submission
Extract data from web pages
Web page and data selectors → Scraped text, links, or structured data
Click buttons and links on web pages
Target element selectors or coordinates → Triggered click actions and page responses
Take screenshots of web pages
Web page or specific page elements → Image files of browser content
Automate login sequences
Username, password, and login form elements → Authenticated browser session
Monitor web page changes
Target web page and monitoring criteria → Alerts or data when page content changes
Tags
Tech Stack
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.