- Install the Pipedream CLI
- Create a Connect project
- Set up a Pipedream OAuth client to authenticate API requests
- Run the SDK playground locally to explore the Pipedream SDK and pre-built tools
Set up Connect with the CLI
1
Install the Pipedream CLI
- macOS
- Linux
- Windows
Using Homebrew:From source:
brew tap pipedreamhq/pd-cli
brew install pipedreamhq/pd-cli/pipedream
curl https://cli.pipedream.com/install | sh
curl https://cli.pipedream.com/install | sh
Download the Windows build, unzip it, and add
pd.exe to your PATH.pd --version
2
Log in to your account
Authenticate with your Pipedream account:This opens your browser to complete sign in.
pd login
3
Initialize a Connect project
Run the Connect initialization command:This command walks you through:
pd init connect
- Creating a new project (or selecting an existing one)
- Setting up a Pipedream OAuth client to authenticate requests to the Connect API
- Choosing one of the available demo apps to get started:
- Next.js app for managed auth : a basic Next.js application demonstrating how your users can connect their accounts
- SDK Playground : an interactive playground that highlights both the frontend
connect-reactand backend@pipedream/sdklibraries for integrating 10,000+ pre-built triggers and actions - MCP Chat App : an AI chat application powered by Pipedream’s MCP server
The CLI automatically creates a
.env file with your OAuth client credentials and project ID. Keep this file secure and don’t commit it to version control.4
Start your chosen demo app
After initialization, start the demo app you selected:Open http://localhost:3000 to see your app running.
cd your-project-name
npm run dev
- Next.js with Managed Auth
- React SDK Playground
- MCP Chat App
This demo showcases:
- User authentication - Sign in flows for your end users
- Account connections - Connect user accounts to + apps
- Basic integration patterns - Common use cases and implementation examples
This interactive playground demonstrates:
- Component browser - Explore 10,000+ pre-built API operations (triggers and actions)
- Managed auth - Connect your users’ accounts to + apps
- Component configuration - Configure and test components with real API data
- Live execution - Run actions and see the results in real-time
This AI chat application features:
- Model Context Protocol (MCP) - Pipedream’s MCP integration for AI agents
- Integrated tools - All + apps available as chat tools
- Live demo - See the production version at chat.pipedream.com
- AI-powered workflows - Execute actions through natural language commands
5
Run the demo app
Follow the instructions in the CLI to set up your local environment and run the app.
Next steps
- Send custom requests with the Connect Proxy
- Explore the Connect API
- Learn more about Pipedream MCP