🧠 Simple MCP Server with Node.js

What is 🧠 Simple MCP Server with Node.js?

A prototype MCP (Model Context Protocol) server built with Node.js and Express that interacts with the OpenAI API to simulate a context‑aware assistant, such as one that helps write stories. It allows structuring context with system, user, steps, and resources, and modifying that context via API to affect LLM responses.

How to use 🧠 Simple MCP Server with Node.js?

Clone the repository, install dependencies with npm install, create a .env file with your OpenAI API key, and create a context.json file with initial context. Then create server.js and openai.js as provided, start the server with node server.js, and use curl commands to get context, generate a plot, or add characters.

Key features of 🧠 Simple MCP Server with Node.js

• Structure context with system, user, steps, and resources
• Generate story plots via OpenAI API using current context
• Add characters dynamically through an API endpoint
• Modify context to see how it affects LLM responses

Use cases of 🧠 Simple MCP Server with Node.js

• Prototyping an LLM‑powered story writing assistant
• Experimenting with structured context for LLM interactions
• Teaching how MCP can be implemented with Node.js and Express

FAQ from 🧠 Simple MCP Server with Node.js

What dependencies are required?

Node.js v18+, npm, and an OpenAI API key.

How does the server store context?

Context is stored in a local context.json file, read from and written to by the server via fs.readFileSync and fs.writeFileSync.

What API endpoints are available?

GET /context returns the current context; POST /step/plot generates a plot using OpenAI; POST /update/characters adds a new character.

What transport does the server use?

The server runs on HTTP via Express, listening on port 3000 by default.

Is there any authentication?

The server itself has no authentication; the OpenAI API key is stored in the .env file and used only when calling the OpenAI API.