Sourcesyncai MCP

🚀 SourceSync.ai MCP Server Configuration Guide

This guide provides a detailed introduction to the installation, configuration, and usage of the SourceSync.ai MCP server, helping you quickly set up and integrate the server into your project.

🚀 Quick Start

Installation Requirements

• Node.js version >= 14.0.0
• npm version >= 6.0.0
• A directory with write permissions for storing data and logs

Download the Source Code

git clone https://github.com/sourcesyncai/mcp.git
cd mcp

Install Dependencies

npm install

Start the Server

node dist/index.js

📦 Installation

Environment Variable Configuration

Edit the config/default.json file and set the following parameters:

• PORT: The port on which the service listens. The default value is 3000.
• BASE_URL: The base path of the service. The default value is /api.
• DEBUG: Whether to enable debug mode. The default value is false.

Log Files

Log files are stored in log/output.log by default. You can modify the log output location and format by editing config/logger.json.

💻 Usage Examples

Document Management Tools

Get Documents

{
  "name": "getDocuments",
  "arguments": {
    "namespaceId": "your_namespace_id",
    "query": "*",
    "filter": {},
    "sort": {},
    "page": 1,
    "perPage": 10,
    "tenantId": "tenant_XXX"
  }
}

Create a Document

{
  "name": "createDocument",
  "arguments": {
    "namespaceId": "your_namespace_id", 
    "title": "Example Document",
    "content": "This is an example document.",
    "tenantId": "tenant_XXX"
  }
}

Update a Document

{
  "name": "updateDocument",
  "arguments": {
    "namespaceId": "your_namespace_id", 
    "documentId": "doc_123456789",
    "title": "Updated Document",
    "content": "This is an updated example document.",
    "tenantId": "tenant_XXX"
  }
}

Data Source Synchronization

Create a Data Source Connection

{
  "name": "create_connection", 
  "arguments": {
    "tenantId": "tenant_XXX",
    "namespaceId": "namespace_XXX",
    "name": "My Connection",
    "connector": "GOOGLE_DRIVE",
    "clientRedirectUrl": "https://your-app.com/callback"
  }
}

Synchronize Data

{
  "name": "sourceSync", 
  "arguments": {
    "tenantId": "tenant_XXX",
    "namespaceId": "namespace_XXX",
    "sourceConfig": {
      "connector": "GOOGLE_DRIVE",
      "config": {
        "path": "/my/drive/path"
      }
    }
  }
}

Document Content Retrieval Workflow

1. First, use getDocuments and set includeConfig.parsedTextFileUrl: true to get the document and its content URL.
2. Extract the URL from the document response.
3. Use fetchUrlContent to get the actual content:

{
  "name": "fetchUrlContent",
  "arguments": {
    "url": "https://example.com"
  }
}

📚 Documentation

Here are some example prompts that can be used with Claude or Cursor (after configuring the MCP server):

• "Search my SourceSync knowledge base for information on machine learning."
• "Ingest this article into my SourceSync knowledge base: [URL]"
• "Create a new namespace for my project documentation."
• "List all documents in my SourceSync namespace."
• "Get the text content of the [document_id] document in my SourceSync namespace."

🔧 Technical Details

Connection Issues

If you encounter issues connecting to the SourceSync.ai MCP server:

1. Verify Paths: Ensure that all paths in the configuration are absolute paths, not relative paths.

2. Check Permissions: Ensure that the server files have execution permissions (chmod +x dist/index.js).

3. Enable Developer Mode: Enable developer mode in the Claude desktop and check the MCP log files.

4. Test the Server: Run the server directly from the command line:

   node /path/to/sourcesyncai-mcp/dist/index.js
   

5. Restart the AI Client: After making changes, completely restart the Claude desktop or Cursor.

6. Check Environment Variables: Ensure that all required environment variables are correctly set.

Log Files

Logs are stored in log/output.log by default. You can modify the log output location and format by editing config/logger.json.

💻 Code Examples

Create an MCP Instance

const MCP = require('sourcesyncai-mcp');
const mcp = new MCP({
  port: 3000,
  baseUrl: '/api',
  debug: true
});
mcp.start();

Send a Request

async function sendRequest() {
  try {
    const response = await fetch('http://localhost:3000/api/document', {
      method: 'POST',
      headers: {
        'Content-Type': 'application/json'
      },
      body: JSON.stringify({
        "name": "createDocument",
        "arguments": {
          "namespaceId": "your_namespace_id", 
          "title": "New Document",
          "content": "This is a new document.",
          "tenantId": "tenant_XXX"
        }
      })
    });
    const data = await response.json();
    console.log(data);
  } catch (error) {
    console.error(error);
  }
}
sendRequest();

📄 Project Links

• SourceSync.ai GitHub Repository
• Official Documentation

By following the above steps, you can successfully configure and run the SourceSync.ai MCP server and integrate it into your project.