kollektiv
Kollektiv MCP enables you to chat with and query your own documents directly from IDEs and MCP clients. Private, secure, and integrated into your favorite code editor
claude mcp add --transport stdio alexander-zuev-kollektiv-mcp npx -y mcp-remote https://mcp.thekollektiv.ai/mcp
How to use
Kollektiv MCP provides a client-friendly way to connect to a remote knowledge base via the MCP protocol. This deprecated server exposes an easier onboarding flow by using the mcp-remote package through npx, allowing editor and IDE clients to talk to Kollektiv without hosting your own MCP backend. To use it, configure your MCP client (Cursor, Windsurf, Claude Desktop, VS Code, Cline, etc.) with the provided JSON configuration. Once connected, you’ll authenticate if required and gain access to Kollektiv’s knowledge-base features through your preferred client. The setup emphasizes a remote MCP endpoint at mcp.thekollektiv.ai/mcp and relies on the mcp-remote package to bridge your client to that endpoint.
How to install
Prerequisites:
- Node.js and npm (or pnpm) installed on your system
- Internet access to fetch packages from npm
Installation and setup steps:
- Ensure Node.js is installed. You can verify with:
node -v
npm -v
- You don’t install a local MCP server binary for this approach; instead you use npx to fetch mcp-remote on demand. Ensure npx is available (comes with npm):
npx --version
- Use the provided configuration in your MCP client. The core connection command uses npx to pull mcp-remote and connect to the remote Kollektiv MCP endpoint:
{
"mcpServers": {
"kollektiv": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://mcp.thekollektiv.ai/mcp"
]
}
}
}
- Depending on your client, save this configuration in the appropriate place:
- Cursor: paste the JSON into the configuration as shown in the README
- Windsurf: Settings > MCP Servers > View raw config, then paste
- VS Code: Settings.json under mcp server configuration
- Cline: cline_mcp_settings.json, add the kolektiv entry
- Start your client and authenticate if prompted. The connection may take a moment to establish; if issues arise, refer to troubleshooting steps in the README.
Additional notes
Important: This MCP server configuration is deprecated and may stop working at any time. The remote endpoint is mcp.thekollektiv.ai/mcp and the active package for connection is mcp-remote (invoked via npx). If you encounter connectivity issues, try restarting the client, waiting a bit, or re-authenticating. Some clients (like Windsurf) may require a restart to establish the connection properly. Ensure you have network access to the endpoint and that your editor/IDE supports the standard MCP JSON format. Since this server is deprecated, consider migrating to the recommended replacement or updated hosting provided by Kollektiv.
Environment considerations:
- No special environment variables are required for the basic connection, but some clients may need their own auth tokens or login flows.
- If you manage multiple MCP servers, you can extend the mcpServers object with additional entries following the same pattern.
Related MCP Servers
mcp-graphql
Model Context Protocol server for GraphQL
anytype
An MCP server enabling AI assistants to interact with Anytype - your encrypted, local and collaborative wiki - to organize objects, lists, and more through natural language.
mcp
MCP server for the Cloudflare API
mcp -circleci
A specialized server implementation for the Model Context Protocol (MCP) designed to integrate with CircleCI's development workflow. This project serves as a bridge between CircleCI's infrastructure and the Model Context Protocol, enabling enhanced AI-powered development experiences.
one-search
🚀 OneSearch MCP Server: Web Search & Scraper & Extract, Support agent-browser, SearXNG, Tavily, DuckDuckGo, Bing, etc.
aica
aica(AI Code Analyzer) reviews your code using AI. Supports CLI and GitHub Actions.