mcpintegrationstutorial

Connect Scavio to Any AI Assistant with MCP

Scavio now supports the Model Context Protocol. Connect to Claude, Cursor, Windsurf, VS Code, ChatGPT, and any MCP-compatible client with a single config -- no code required.

6 min read

AI assistants are most useful when they can reach beyond their training data. If you ask Claude to compare headphone prices on Amazon, or pull the transcript from a YouTube video, it needs a way to call those APIs in real time. Until now, connecting Scavio to your AI assistant meant writing code -- a LangChain tool, an OpenClaw skill, or a direct API call.

Today, Scavio supports the Model Context Protocol (MCP). This means you can connect Scavio to Claude, Cursor, Windsurf, VS Code, ChatGPT, and any other MCP-compatible client with a single config -- no code required.

What Is MCP?

MCP is an open standard created by Anthropic for connecting AI assistants to external tools and data sources. Instead of building custom integrations for every AI client, a service publishes one MCP server and every compatible client can connect to it. Think of it as a USB-C port for AI tools: one plug, every device.

Scavio's MCP server runs at https://mcp.scavio.dev/mcp and uses HTTP transport, so there is nothing to install locally. Your AI assistant connects directly over the network.

What Your AI Assistant Gets

Once connected, your assistant has access to 9 tools covering four platforms:

ToolWhat it does
search_googleWeb search with knowledge graph, People Also Ask, news, images, maps
search_amazonSearch Amazon products with price and sort filters
get_amazon_productFull product details by ASIN
search_walmartSearch Walmart products with delivery and price filters
get_walmart_productFull product details by product ID
search_youtubeSearch videos, channels, and playlists
get_youtube_metadataVideo title, views, likes, duration, channel info
get_youtube_transcriptTimestamped video transcripts
get_usageCheck credit balance and plan details

Quick Setup: Claude Code

One command:

Bash
claude mcp add --transport http scavio https://mcp.scavio.dev/mcp \
  --header "x-api-key: YOUR_SCAVIO_API_KEY"

Verify with claude mcp list. Your Claude Code session now has access to all 9 tools.

Quick Setup: Cursor

Go to Settings > Tools & MCP > + Add New MCP Server, select HTTP, and enter:

  • Name: scavio
  • URL: https://mcp.scavio.dev/mcp
  • Headers: x-api-key: YOUR_SCAVIO_API_KEY

Or add this to .cursor/mcp.json:

JSON
{
  "mcpServers": {
    "scavio": {
      "type": "http",
      "url": "https://mcp.scavio.dev/mcp",
      "headers": {
        "x-api-key": "YOUR_SCAVIO_API_KEY"
      }
    }
  }
}

Quick Setup: VS Code

Create .vscode/mcp.json in your project root:

JSON
{
  "servers": {
    "scavio": {
      "type": "http",
      "url": "https://mcp.scavio.dev/mcp",
      "headers": {
        "x-api-key": "YOUR_SCAVIO_API_KEY"
      }
    }
  }
}

Open the Command Palette and run MCP: List Servers to confirm.

Works Everywhere

The same config pattern works with Claude Desktop, Windsurf, ChatGPT, Cline, Zed, and any other MCP-compatible client. The universal config is just three fields:

JSON
{
  "type": "http",
  "url": "https://mcp.scavio.dev/mcp",
  "headers": {
    "x-api-key": "YOUR_SCAVIO_API_KEY"
  }
}

For step-by-step instructions for every supported client, see the full MCP integration guide.

Get Started

Sign up at dashboard.scavio.dev to get your API key. The free tier includes 500 credits per month -- no credit card required. Each MCP tool call costs 1 credit (light mode) or 2 credits (full mode with knowledge graphs, People Also Ask, and more).

Read the MCP integration docs for the complete setup guide, available tools reference, and troubleshooting tips.