1xn-vmcp

The 1xn-vmcp server provides a vMCP (virtual Model Context Protocol) layer that aggregates multiple MCP servers into a single, configurable workspace. It enables you to compose diverse MCPs into a unified virtual MCP, extend their tooling, and manage authentication and usage statistics across clients. Through the vMCP interface you can add remote MCP servers, override tool names and descriptions, create programmable prompts, and bring in custom resources or knowledge bases to a single endpoint. This makes it possible to route queries through a curated set of tools and resources, enabling complex workflows and tool chaining from Claude, ChatGPT, Gemini, VSCode, Cursor, and other agents.

To use the server, install it via uvx, uv, pip, or Docker as shown in the installation steps, then run the VMCP server. Once running, access the UI at the default host and port (http://localhost:8000) to create a vMCP, add MCP connectors, and configure tools and prompts. The Tools tab lets you enable/disable or rename tools from connected MCP servers, while the Prompts tab allows programmable prompts that can invoke other tools or resources, enabling user-controlled tool chaining within your virtual MCP. You can also add custom resources or documents to enrich your knowledge base alongside the connected MCPs.

Prerequisites:

• Python 3.10 to 3.13
• uv (or uvx) for quick trial
• Optional: Docker for container-based runs

Installation options:

Option 1: Using uvx (recommended)

uvx --from 1xn-vmcp@latest vmcp run

This installs and starts the vMCP server, exposing it at http://localhost:8000.

Option 2: Using uv (via uv by spreading packages)

uv install 1xn-vmcp
uv run vmcp

Option 3: Using pip
```bash
pip install 1xn-vmcp
vmcp run

Option 4: Using Docker

docker pull onexn/vmcp:latest
docker run -it onexn/vmcp:latest vmcp run

After starting, the server will be accessible at http://localhost:8000 and ready to accept vMCP configurations.

Tips and notes:

• The vMCP platform supports composing multiple MCP servers into a single virtual MCP. Use the UI to add connectors, then configure tools, prompts, and resources per vMCP.
• You can override tool names and descriptions from upstream MCPs to tailor them to your workflow.
• Ensure environment variables (auth tokens, endpoints) are provided in the vMCP editor when connecting custom servers.
• Docker deployments are convenient for quick trials; be aware of port mappings if you run behind a reverse proxy.
• If you encounter connection issues, verify that the MCP servers you added are reachable and that authentication headers are correctly configured in the vMCP editor.