Quick Agent Start

A Model Context Protocol (MCP) server that enables Claude Desktop, VSCode, Cursor and other AI agents to upload files and query Synvo AI's powerful document analysis capabilities directly from your development environment.

What Can This MCP Server Do?

This server adds two powerful capabilities to your AI agent:

🚀 Upload & Analyze Files

Upload any document (PDF, DOCX, TXT, images, etc.) to Synvo AI for content extraction and analysis

💬 Query File Contents

Ask questions about uploaded documents and get AI-powered insights with multi-turn conversations

Quick Setup (5 Minutes)

Step 1: Installation

# Install from npm (recommended)
npm install -g @synvo_ai/mcp-server
# Or clone and build locally
git clone https://github.com/synvoai/mcp-server.git
cd mcp-server
npm install
npm run build

Step 2: Get Your API Key

Get your free API key from Synvo Dashboard.

  1. Log into your Synvo dashboard
  2. Navigate to API Keys section
  3. Click "Create New Key"
  4. Copy and save the key immediately - you won't see it again!

Step 3: Configure Your AI Client

For macOS:

  1. Open: ~/Library/Application Support/Claude/claude_desktop_config.json

For Windows:

  1. Open: %APPDATA%\Claude\claude_desktop_config.json

  2. Add this configuration:

{
  "mcpServers": {
    "synvo": {
      "command": "npx",
      "args": ["-y", "@synvo_ai/mcp-server"],
      "env": {
        "SYNVO_API_KEY": "your-api-key-here"
      }
    }
  }
}
  1. Restart Claude Desktop

Add to your MCP settings:

{
  "mcp.servers": {
    "synvo": {
      "command": "npx",
      "args": ["-y", "@synvo_ai/mcp-server"],
      "env": {
        "SYNVO_API_KEY": "your-api-key-here"
      }
    }
  }
}

Usage Examples

Once configured, you can interact with your AI agent using natural language:

Example 1: Analyze a Document

💬 User:

"Upload and analyze this PDF file: /Users/lucy/Documents/report.pdf"

🤖 AI Agent:

✅ File uploaded successfully!

File ID: abc123...

The file is now available for analysis.

Example 2: Query File Contents

💬 User:

"What are the main findings in that report?"

🤖 AI Agent:

📝 AI Response:

The report identifies three main findings:

  1. Revenue increased by 25% year-over-year
  2. Customer satisfaction improved to 92%
  3. Operating costs decreased by 15%

💬 Conversation ID: conv_xyz789

(Use this ID to continue the conversation)

Example 3: Multi-Turn Conversation

💬 User:

"Can you give me more details about the cost reduction?"

🤖 AI Agent:

📝 AI Response:

The 15% cost reduction was achieved through:

  • Automation of manual processes (8% savings)
  • Renegotiated supplier contracts (5% savings)
  • Energy efficiency improvements (2% savings)

Available Tools

1. synvo_upload_file

Upload and analyze documents to make them available for AI queries.

Parameters:

  • file_path (required): Absolute path to the file
  • build_memory (optional): Build AI memory from content (default: true)
{
  "file_path": "/Users/lucy/Documents/contract.pdf",
  "build_memory": true
}

2. synvo_query_ai

Ask questions about uploaded files or have AI conversations.

Parameters:

  • message (required): Your question or instruction
  • conversation_id (optional): Continue a previous conversation
  • stream (optional): Stream the response (default: false)
{
  "message": "Summarize the key points from the contract",
  "conversation_id": "conv_abc123"
}

Common Use Cases

📄 Document Analysis

  • "Upload this research paper and extract the methodology section"
  • "Analyze this contract and highlight any unusual clauses"
  • "Summarize the key findings from this financial report"

🔄 Multi-Document Comparison

  • "Upload these three proposals and compare their pricing"
  • "What are the differences between document A and document B?"

🔍 Information Extraction

  • "Extract all dates and deadlines from this project plan"
  • "List all the action items mentioned in this meeting transcript"
  • "Find all references to 'budget' in these documents"

🌐 Translation & Transformation

  • "Translate the content of this document to Spanish"
  • "Convert this technical document into simple language"
  • "Create a bullet-point summary of this report"

Supported File Types

📄 Documents

PDF, DOCX, DOC, TXT, RTF

📊 Spreadsheets

XLSX, XLS, CSV

🖼️ Images

PNG, JPG, JPEG, GIF (with OCR)

📽️ Presentations

PPTX, PPT

💻 Code

JS, PY, TS, Java, etc.

Troubleshooting

Server Not Starting

  • Check if Node.js 18+ is installed: node --version
  • Verify the build completed: npm run build
  • Check logs in Claude Desktop: Menu → Settings → Developer

Upload Failures

  • Ensure file path is absolute (not relative)
  • Verify file exists and is readable
  • Check file size (max 50MB recommended)
  • Confirm API key is valid

Query Not Working

  • Upload a file first before querying
  • Check if conversation_id is valid
  • Verify network connectivity to api.synvo.ai

Getting Help

Advanced Configuration

Using a Local Build Instead of NPM

If you prefer to use a local build:

{
  "mcpServers": {
    "synvo": {
      "command": "node",
      "args": ["C:\\path\\to\\synvo-mcp\\dist\\index.js"],
      "env": {
        "SYNVO_API_KEY": "your-api-key-here"
      }
    }
  }
}
{
  "mcpServers": {
    "synvo": {
      "command": "node",
      "args": ["/path/to/synvo-mcp/dist/index.js"],
      "env": {
        "SYNVO_API_KEY": "your-api-key-here"
      }
    }
  }
}

Environment Variables

You can also set the API key globally:

setx SYNVO_API_KEY "your-api-key-here"

Then simplify your config:

{
  "mcpServers": {
    "synvo": {
      "command": "npx",
      "args": ["-y", "@synvo_ai/mcp-server"]
    }
  }
}
export SYNVO_API_KEY="your-api-key-here"

Then simplify your config:

{
  "mcpServers": {
    "synvo": {
      "command": "npx",
      "args": ["-y", "@synvo_ai/mcp-server"]
    }
  }
}

What's Next?


Need Help? Contact support@synvo.ai or visit https://synvo.ai/support