mcp
MCP server from supavec/mcp-server
claude mcp add --transport stdio supavec-mcp-server npx -y @supavec/mcp-server --api-key <your_api_key>
How to use
The Supavec MCP Server exposes an API that lets AI assistants fetch relevant content and embeddings from your Supavec files. It integrates with MCP-compatible tools such as Cursor, VS Code Copilot, Claude, and other MCP-enabled assistants, enabling you to search your Supavec storage and retrieve embedding-based results for informed responses. Core features include fetch-embeddings for targeted retrieval within a file and list-user-files to enumerate your uploaded documents. Configuration is designed to be straightforward: point your MCP client to the server and supply your API key (either via a command-line argument or environment variable). The server supports both direct CLI usage and integration through popular editors and assistants by adding an mcp.json configuration snippet in the respective tool’s setup.
Available tools:
- fetch-embeddings: Retrieve embeddings for a specific file by its ID and run a query to locate relevant passages or sections within that file.
- list-user-files: List all files uploaded to your Supavec account, with optional pagination and ordering. Responses include file metadata such as ID, name, type, creation date, and team association. Use these to discover which files to query with fetch-embeddings.
How to install
Prerequisites:
- Node.js 16.0.0 or higher
- npm (comes with Node.js)
Install and run via npx (recommended):
- Ensure you have an API key from Supavec.
- Run the server directly without global installation:
npx @supavec/mcp-server --api-key your_api_key
Optional: Install globally for easier reuse:
npm install -g @supavec/mcp-server
supavec-mcp --api-key your_api_key
Environment variables alternative:
- SUPAVEC_API_KEY: Set this in your environment and omit the --api-key argument in your mcp.json/config. Example:
export SUPAVEC_API_KEY=your_api_key
Then configure with:
{
"mcpServers": {
"supavec": {
"command": "npx",
"args": ["-y", "@supavec/mcp-server"]
}
}
}
If you build from source, typical flow is:
git clone https://github.com/supavec/mcp-server.git
cd mcp-server
npm install
npm run build
Then run via node if you build a local distribution:
node build/index.js --help
Additional notes
Tips and notes:
- The API key has the highest priority when supplied as a command-line argument. If not provided, the server will fall back to the SUPAVEC_API_KEY environment variable.
- If you encounter authentication or 401 errors, verify that your API key is active and has permissions to access your Supavec files.
- Use the debugging option (e.g., DEBUG=1) to get more logs for troubleshooting.
- The server supports both per-file embedding queries and broader file listing; combine list-user-files with fetch-embeddings to iteratively locate relevant content.
- When integrating with editors or assistants, ensure your MCP JSON points to the same server and uses consistent API key handling.
Related MCP Servers
CanvasMCPClient
Canvas MCP Client is an open-source, self-hostable dashboard application built around an infinite, zoomable, and pannable canvas. It provides a unified interface for interacting with multiple MCP (Model Context Protocol) servers through a flexible, widget-based system.
docmole
Dig through any documentation with AI - MCP server for Claude, Cursor, and other AI assistants
obsidian
MCP server for Obsidian vault management - enables Claude and other AI assistants to read, write, search, and organize your notes
GameMaker
GameMaker MCP server for Cursor - Build GM projects with AI
xgmem
Global Memory MCP server, that manage all projects data.
mcp-turso
MCP server for interacting with Turso-hosted LibSQL databases