Activepieces
VerifiedOpen source Zapier alternative exposing 280+ pieces as MCP servers.
What is the Activepieces MCP server?
Activepieces provides an intuitive no-code builder for workflows while allowing developers to create custom pieces in TypeScript that publish to npm. These pieces integrate services like Google Sheets, OpenAI, and Discord and become usable directly by AI agents via the MCP protocol.
The platform emphasizes security through self-hosting, human-in-the-loop approvals, and versioned flows. It includes native AI pieces and an SDK for building agents inside the automation builder.
Install & connect
Set up this server, then add it to your MCP client.
Full setup instructions are in the GitHub repository.
Example prompts
Once connected, try asking your AI client:
Security & permissions
Runs locally via stdio as a self-hosted application; requires access to any API keys or credentials configured inside the pieces and full control over the host machine's network.
What you can do with Activepieces
AI Agent Integrations
Expose Activepieces pieces as MCP tools so LLMs can trigger automations across 200+ services without writing custom code.
Self-Hosted Workflow Automation
Run private, network-gapped automations that combine loops, branches, code execution with npm, and AI actions.
Community Piece Development
Build and publish TypeScript pieces that instantly become MCP servers available to AI clients after npm release.
How to use Activepieces
- 1Install and run Activepieces locally or via Docker following the deploy docs.
- 2Create or select pieces from the 280+ available integrations.
- 3Enable MCP exposure for the desired pieces in the platform settings.
- 4Configure the MCP server command in Claude Desktop, Cursor, or Windsurf.
- 5Start chatting with your AI client to invoke the exposed pieces.
Activepieces: pros & cons
Pros
- +280+ pieces automatically become MCP servers without extra work
- +Strong TypeScript developer experience with hot reloading
- +Self-hosted and open source under MIT license
- +Active community contributing 60% of pieces
Cons
- –MCP tools are only as granular as the underlying pieces
- –Requires self-hosting setup and maintenance
- –No pre-defined MCP tool schemas documented in the README
Frequently asked questions
Over 280 pieces are automatically exposed as MCP servers once contributed.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…
Similar MCP servers
Other ai & knowledge connectors worth a look.
