MCP 服务器

Peliqan MCP Server

Enables AI assistants to interact with the Peliqan platform, providing access to hundreds of data connectors for reading and writing data in business applications.

README

Peliqan MCP Server

For more info, visit https://peliqan.io/mcp-docs first.

[!NOTE]
This repo is only useful if you want to run a local MCP Server. For most use cases, the above link is all you need to get started in ChatGPT, Claude.ai, Cursor etc.

Peliqan Local MCP Server

This repo provides a local MCP Server, that you can install on your computer. It communicates with the Peliqan platform via a "Proxy" API endpoint.

Configuration

Set up the connection to your Peliqan acount.

Create a configuration file named peliqan_config.json in your home folder:

{
    "account_id": 1234,
    "api_token": "xxx"
}

Or add the environment variables:

account_id
api_token

Installation

pip install mcp-server-peliqan

Usage with Claude Desktop

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "peliqan": {
      "command": "python",
      "args": [
        "-m",
        "mcp_server_peliqan"
      ],
      "env": {
        "peliqan_account_id": "1234",
        "peliqan_api_token": "xxx"
      }
    }
  }
}

Location on Windows: $env:AppData\Claude\claude_desktop_config.json

Location on Mac: ~/Library/Application\ Support/Claude/claude_desktop_config.json

Setup in Peliqan

Add an API handler script, that exposes one or more MCP methods (tools). See the "MCP" templates in Peliqan to get started.
Add a custom API endpoint in your Peliqan account with route /mcp and method POST, and link it to the API handler script.

Supported sources

All Peliqan connectors can be used in your MCP Server to fetch data and to perform writeback, for example adding/updating resources in your SaaS business applications (contacts, invoices, deals etc.).

Peliqan connectors: