Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Nullplatform API MCPlist all my current deployments"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Nullplatform API MCP
This project exposes the Nullplatform OpenAPI an MCP so that large language models (LLMs) can discover, understand, and interact with the nullplatform's API endpoints.
It uses the official OpenAPI spec from https://docs.nullplatform.com/ and automatically registers each endpoint as a function that the LLM can invoke.
✨ What It Does
Loads the Nullplatform OpenAPI spec.
Registers all available endpoints (GET, POST, etc.) as discoverable callable functions.
This mcp uses as base software @nullplatform/meta-mcp-proxy which allows you to do an in-place RAG for scenarios where you have a big amount of tools
Exposes those functions via MCP so the LLM can:
List, get, create, and modify deployments, scopes, services, and more.
Understand required parameters and request bodies.
Follow usage guidance like retrieving the
nrnbefore modifying resources.
🧱 Installation
Using npx
Edit your file ~/Library/Application Support/Claude/claude_desktop_config.json and add:
🐳 Using Docker
Build the image:
Configure Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
Or for Claude Code CLI (.mcp.json in your project):