Activepieces

What is Activepieces about?

Activepieces delivers an all‑in‑one automation engine where users design flows with triggers, actions, loops, and branches. The system is built around pieces, npm packages written in TypeScript that expose functionality as MCP servers, making them callable from LLMs such as Claude Desktop, Cursor, or Windsurf.

How to use Activepieces?

1. Deploy – Follow the official deployment guide (Docker, cloud VM, or local Node.js).
2. Create a flow – Use the web UI to drag‑and‑drop triggers (e.g., HTTP, schedule, human input) and actions (e.g., Google Sheets, OpenAI, custom code).
3. Add custom pieces – Build a new piece with the TypeScript framework, publish it to npm, and it automatically appears as an MCP endpoint inside the builder.
4. Run flows – Execute instantly, schedule, or invoke via API/MCP from any LLM‑powered assistant.

Key features of Activepieces

• Extensible type‑safe pieces (280+ community pieces, hot‑reloading for local development).
• AI‑first design – Native AI pieces, Copilot assistance, and direct LLM integration via MCP.
• Human‑in‑the‑loop – Chat and form triggers, approval steps, delayed execution.
• Full version control – Every flow is versioned, supporting rollbacks and audits.
• Enterprise‑ready – Self‑hosted, network‑gapped deployment, customizable branding and access control.
• Open ecosystem – All source code lives on GitHub, pieces are published to npm, and 60 % are community contributions.

Use cases of Activepieces

• Automated reporting – Pull data from databases, generate PDFs, and email results on a schedule.
• Customer support bots – Combine chat triggers with AI pieces to fetch ticket data, suggest responses, and log interactions.
• HR onboarding – Chain form inputs, document generation, and Slack notifications into a single flow.
• Data enrichment – Trigger on new rows in Google Sheets, call external APIs (e.g., OpenAI, CRM), and write back enriched data.
• Custom LLM agents – Expose internal services as MCP endpoints that LLMs can call during reasoning or tool‑use phases.

FAQ from the Activepieces community

• Do I need to code to create a flow? No. The visual builder is intended for non‑technical users; custom code pieces are optional for developers.
• Can I run Activepieces on a private network? Yes. The platform is fully self‑hostable and can be placed behind firewalls or VPNs.
• How are pieces versioned? Pieces follow standard npm versioning; the platform fetches the specified version when a flow is published.
• Is there a free tier? The Community Edition is MIT‑licensed and free to use; an Enterprise edition adds commercial licensing and advanced features.
• What languages are supported for triggers? Triggers are language‑agnostic; they can be HTTP, schedule, or UI‑based (chat/form) and invoke any piece regardless of implementation language.

🤯 Welcome to Activepieces

All-in-one AI automation designed to be extensible through a type-safe pieces framework written in TypeScript. When you contribute pieces to Activepieces they become automatically available as MCP servers that you can use with LLMs through Claude Desktop, Cursor or Windsurf!

🔥 Why Activepieces is Different:

• 💖 Loved by Everyone: Intuitive interface and great experience for both technical and non-technical users with a quick learning curve.

• 🌐 Open Ecosystem: All pieces are open source and available on npmjs.com, 60% of the pieces are contributed by the community.

• 🛠️ Largest open source MCP toolkit: All our pieces (280+) are available as MCP that you can use with LLMs on Claude Desktop, Cursor or Windsurf.

• 🛠️ Pieces are written in Typescript: Pieces are npm packages in TypeScript, offering full customization with the best developer experience, including hot reloading for local piece development on your machine. 😎

• 🤖 AI-First: Native AI pieces let you experiment with various providers, or create your own agents using our AI SDK, and there is Copilot to help you build flows inside the builder.

• 🏢 Enterprise-Ready: Developers set up the tools, and anyone in the organization can use the no-code builder. Full customization from branding to control.

• 🔒 Secure by Design: Self-hosted and network-gapped for maximum security and control over your data.

• 🧠 Human in the Loop: Delay execution for a period of time or require approval. These are just pieces built on top of the piece framework, and you can build many pieces like that. 🎨

• 💻 Human Input Interfaces: Built-in support for human input triggers like "Chat Interface" 💬 and "Form Interface" 📝

🛠️ Builder Features:

• Loops
• Branches
• Auto Retries
• HTTP
• Code with NPM
• ASK AI in Code Piece (Non technical user can clean data without knowing to code)
• Flows are fully versioned.
• Languages Translations
• Customizable Templates
• 200+ Pieces, check https://www.activepieces.com/pieces

We release updates frequently. Check the product changelog for the latest features.

🔌 Create Your Own Piece

Activepieces supports integrations with Google Sheets, OpenAI, Discord, RSS, and over 200 other services. Check out the full list of supported integrations, which is constantly expanding thanks to our community's contributions.

As an open ecosystem, all integration source code is accessible in our repository. These integrations are versioned and published directly to npmjs.com upon contribution.

You can easily create your own integration using our TypeScript framework. For detailed instructions, please refer to our Contributor's Guide.

License

Activepieces' Community Edition is released as open source under the MIT license and enterprise features are released under Commercial License

Read more about the feature comparison here https://www.activepieces.com/docs/about/editions

💭 Join Our Community

🌐 Contributions

We welcome contributions big or small and in different directions. The best way to do this is to check this document and we are always up to talk on our Discord Server.

📚 Translations

Not into coding but still interested in contributing? Come join our Discord and visit https://www.activepieces.com/docs/about/i18n for more information.

🦫 Contributors

This project follows the all-contributors specification. Contributions of any kind are welcome!