base on ⚙️ Create and run workflows (RPA 2.0) <picture> <img alt="Workflow Use logo - a product by Browser Use." src="./static/workflow-use.png" width="full"> </picture> <br /> <h1 align="center">Deterministic, Self Healing Workflows (RPA 2.0)</h1> [](https://github.com/browser-use/workflow-use/stargazers) [](https://link.browser-use.com/discord) [](https://cloud.browser-use.com) [](https://x.com/gregpr07) [](https://x.com/mamagnus00) ⚙️ **Workflow Use** is the easiest way to create and execute deterministic workflows with variables which fallback to [Browser Use](https://github.com/browser-use/browser-use) if a step fails. You just _show_ the recorder the workflow, we automatically generate the workflow. ❗ This project is in very early development so we don't recommend using this in production. Lots of things will change and we don't have a release schedule yet. Originally, the project was born out of customer demand to make Browser Use more reliable and deterministic. # Quick start ```bash git clone https://github.com/browser-use/workflow-use ``` ## Build the extension ```bash cd extension && npm install && npm run build ``` ## Setup workflow environment ```bash cd .. && cd workflows uv sync source .venv/bin/activate # for mac / linux playwright install chromium cp .env.example .env # add your OPENAI_API_KEY to the .env file ``` ## Run workflow as tool ```bash python cli.py run-as-tool examples/example.workflow.json --prompt "fill the form with example data" ``` ## Run workflow with predefined variables ```bash python cli.py run-workflow examples/example.workflow.json ``` ## Record your own workflow ```bash python cli.py create-workflow ``` ## See all commands ```bash python cli.py --help ``` # Usage from python Running the workflow files is as simple as: ```python from workflow_use import Workflow workflow = Workflow.load_from_file("example.workflow.json") result = asyncio.run(workflow.run_as_tool("I want to search for 'workflow use'")) ``` ## Launch the GUI The Workflow UI provides a visual interface for managing, viewing, and executing workflows. ### Option 1: Using the CLI command (Recommended) The easiest way to start the GUI is with the built-in CLI command: ```bash cd workflows python cli.py launch-gui ``` This command will: - Start the backend server (FastAPI) - Start the frontend development server - Automatically open http://localhost:5173 in your browser - Capture logs to the `./tmp/logs` directory Press Ctrl+C to stop both servers when you're done. ### Option 2: Start servers separately Alternatively, you can start the servers individually: #### Start the backend server ```bash cd workflows uvicorn backend.api:app --reload ``` #### Start the frontend development server ```bash cd ui npm install npm run dev ``` Once both servers are running, you can access the Workflow GUI at http://localhost:5173 in your browser. The UI allows you to: - Visualize workflows as interactive graphs - Execute workflows with custom input parameters - Monitor workflow execution logs in real-time - Edit workflow metadata and details # Demos ## Workflow Use filling out form instantly https://github.com/user-attachments/assets/cf284e08-8c8c-484a-820a-02c507de11d4 ## Gregor's explanation https://github.com/user-attachments/assets/379e57c7-f03e-4eb9-8184-521377d5c0f9 # Features - 🔁 **Record Once, Reuse Forever**: Record browser interactions once and replay them indefinitely. - ⏳ **Show, don't prompt**: No need to spend hours prompting Browser Use to do the same thing over and over again. - ⚙️ **Structured & Executable Workflows**: Converts recordings into deterministic, fast, and reliable workflows which automatically extract variables from forms. - 🪄 **Human-like Interaction Understanding**: Intelligently filters noise from recordings to create meaningful workflows. - 🔒 **Enterprise-Ready Foundation**: Built for future scalability with features like self-healing and workflow diffs. # Vision and roadmap Show computer what it needs to do once, and it will do it over and over again without any human intervention. ## Workflows - [ ] Nice way to use the `.json` files inside python code - [ ] Improve LLM fallback when step fails (currently really bad) - [ ] Self healing, if it fails automatically agent kicks in and updates the workflow file - [ ] Better support for LLM steps - [ ] Take output from previous steps and use it as input for next steps - [ ] Expose workflows as MCP tools - [ ] Use Browser Use to automatically create workflows from websites ## Developer experience - [ ] Improve CLI - [ ] Improve extension - [ ] Step editor ## Agent - [ ] Allow Browser Use to use the workflows as MCP tools - [ ] Use workflows as website caching layer ", Assign "at most 3 tags" to the expected json: {"id":"14545","tags":[]} "only from the tags list I provide: [{"id":77,"name":"3d"},{"id":89,"name":"agent"},{"id":17,"name":"ai"},{"id":54,"name":"algorithm"},{"id":24,"name":"api"},{"id":44,"name":"authentication"},{"id":3,"name":"aws"},{"id":27,"name":"backend"},{"id":60,"name":"benchmark"},{"id":72,"name":"best-practices"},{"id":39,"name":"bitcoin"},{"id":37,"name":"blockchain"},{"id":1,"name":"blog"},{"id":45,"name":"bundler"},{"id":58,"name":"cache"},{"id":21,"name":"chat"},{"id":49,"name":"cicd"},{"id":4,"name":"cli"},{"id":64,"name":"cloud-native"},{"id":48,"name":"cms"},{"id":61,"name":"compiler"},{"id":68,"name":"containerization"},{"id":92,"name":"crm"},{"id":34,"name":"data"},{"id":47,"name":"database"},{"id":8,"name":"declarative-gui "},{"id":9,"name":"deploy-tool"},{"id":53,"name":"desktop-app"},{"id":6,"name":"dev-exp-lib"},{"id":59,"name":"dev-tool"},{"id":13,"name":"ecommerce"},{"id":26,"name":"editor"},{"id":66,"name":"emulator"},{"id":62,"name":"filesystem"},{"id":80,"name":"finance"},{"id":15,"name":"firmware"},{"id":73,"name":"for-fun"},{"id":2,"name":"framework"},{"id":11,"name":"frontend"},{"id":22,"name":"game"},{"id":81,"name":"game-engine "},{"id":23,"name":"graphql"},{"id":84,"name":"gui"},{"id":91,"name":"http"},{"id":5,"name":"http-client"},{"id":51,"name":"iac"},{"id":30,"name":"ide"},{"id":78,"name":"iot"},{"id":40,"name":"json"},{"id":83,"name":"julian"},{"id":38,"name":"k8s"},{"id":31,"name":"language"},{"id":10,"name":"learning-resource"},{"id":33,"name":"lib"},{"id":41,"name":"linter"},{"id":28,"name":"lms"},{"id":16,"name":"logging"},{"id":76,"name":"low-code"},{"id":90,"name":"message-queue"},{"id":42,"name":"mobile-app"},{"id":18,"name":"monitoring"},{"id":36,"name":"networking"},{"id":7,"name":"node-version"},{"id":55,"name":"nosql"},{"id":57,"name":"observability"},{"id":46,"name":"orm"},{"id":52,"name":"os"},{"id":14,"name":"parser"},{"id":74,"name":"react"},{"id":82,"name":"real-time"},{"id":56,"name":"robot"},{"id":65,"name":"runtime"},{"id":32,"name":"sdk"},{"id":71,"name":"search"},{"id":63,"name":"secrets"},{"id":25,"name":"security"},{"id":85,"name":"server"},{"id":86,"name":"serverless"},{"id":70,"name":"storage"},{"id":75,"name":"system-design"},{"id":79,"name":"terminal"},{"id":29,"name":"testing"},{"id":12,"name":"ui"},{"id":50,"name":"ux"},{"id":88,"name":"video"},{"id":20,"name":"web-app"},{"id":35,"name":"web-server"},{"id":43,"name":"webassembly"},{"id":69,"name":"workflow"},{"id":87,"name":"yaml"}]" returns me the "expected json"