Quick Start for MCP Agents

The Model Context Protocol (MCP) server seamlessly connects AI agents to Synvo AI's powerful multimodal contextual memory system. This integration enables:

• File uploads and processing across multiple formats
• Deep document understanding and analysis (beta)
• User intent recognition (coming soon)
• Automated workflows (coming soon)
• Evidence-based contextual query reasoning

This guide will show you how to integrate Synvo AI with your AI agent using MCP, allowing your agent to upload, analyze, and query files with precise, evidence-grounded understanding.

Overview

The Synvo MCP server extends Claude’s capabilities with two key functions:

🚀 Upload & Analyze Files

Agents can upload documents, images, videos, and audio files to Synvo’s contextual memory, where each file is processed into structured, retrievable context.

🔍 Query with Context

Claude can then ask grounded, multi-modal questions across uploaded content — retrieving accurate, contextualized responses instead of surface-level summaries.

Getting Started

1. Install the MCP Server

npm install -g @synvo/mcp-server

2. Configure Claude Desktop

Add the following to your Claude Desktop configuration:

{
  "mcpServers": {
    "synvo": {
      "command": "synvo-mcp-server",
      "args": ["--api-key", "YOUR_SYNVO_API_KEY"]
    }
  }
}

3. Get Your API Key

1. Visit Synvo Console
2. Create a new API key
3. Copy the key to your configuration

4. Start Using with Claude

Once configured, you can use these commands in Claude:

• Upload a file: "Upload this document to Synvo"
• Query content: "What does this document say about [topic]?"
• Search across files: "Find information about [topic] in my uploaded files"

Example Usage

Upload and Query a Document

User: Upload this PDF to Synvo and then tell me what it says about the company's revenue growth.

Claude: I'll upload the PDF to Synvo and analyze it for you.

[Uploads file to Synvo]

Based on the document, the company shows a 15% year-over-year revenue growth, primarily driven by increased subscription sales in the enterprise segment.

Multi-Modal Analysis

User: Upload this image and video, then compare the information in both.

Claude: I'll upload both files to Synvo and provide a comparative analysis.

[Uploads both files]

The image shows the event poster with basic details, while the video provides additional context about the speaker's background and presentation topics.

Supported File Types

• Images: .png, .jpg, .jpeg, .gif, .webp
• Video: .mp4, .mov, .m4v, .avi, .mkv
• Audio: .mp3, .wav, .m4a, .aac, .flac
• Documents: .pdf, .docx, .ppt, .pptx, .txt, .json, .jsonl, .xlsx

Troubleshooting

Connection Issues

• Verify your API key is correct
• Check that the MCP server is properly installed
• Ensure Claude Desktop is restarted after configuration changes

File Upload Problems

• Check file size limits (typically 100MB per file)
• Verify supported file formats
• Ensure stable internet connection

Query Issues

• Try rephrasing your questions
• Use more specific queries for better results
• Check that files have finished processing before querying

Next Steps

1. API Reference - Complete API documentation
2. File Management - Advanced file operations
3. Query Features - Advanced querying capabilities
4. Authentication - API key management

When AI agents need to understand their user, the Synvo MCP server gives them access to rich, multimodal context — enabling precise, verifiable understanding across every document, image, video, or audio source.