mcpField
O
OpenAI Complete MCP Server

OpenAI Complete MCP Server

Local
github.com/aiamblichusAI & Agents0Updated 84w ago
Repository

About OpenAI Complete MCP Server

OpenAI Complete MCP Server connects MCP clients to OpenAI-compatible text completion APIs, primarily for base models. It offers a single completion tool, without chat completion support.

Key features

  1. 01Single `complete` tool
  2. 02OpenAI-compatible API support
  3. 03Async request handling
  4. 04Timeout fallbacks
  5. 05Request cancellation

Use cases

  1. 01Generating text with a base model
  2. 02Setting completion limits and sampling parameters
  3. 03Cancelling an ongoing completion

MCP support

Exposes a `complete` tool to MCP clients over stdio. The tool accepts a prompt and optional generation parameters.

#mcp-openai-complete#text-completion#openai

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "openai-complete": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "build"
      ]
    }
  }
}
Run commanddocker run -i --rm build

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
ocibuild

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

OpenAI Complete MCP Server

An MCP (Model Context Protocol) server that provides a clean interface for LLMs to use text completion capabilities through the MCP protocol. This server acts as a bridge between an LLM client and any OpenAI's compatible API. The primary use case is for base models, as the server does not provide support for chat completions.

Features

  • Provides a single tool named "complete" for generating text completions
  • Properly handles asynchronous processing to avoid blocking
  • Implements timeout handling with graceful fallbacks
  • Supports cancellation of ongoing requests

Installation

# Clone the repository
git clone <repository-url>
cd mcp-openai-complete

# Install dependencies
pnpm install

# Build the project
pnpm run build

Configuration

The following environment variables are required:

OPENAI_API_KEY=your-hyperbolic-api-key
OPENAI_API_BASE=https://api.hyperbolic.xyz/v1
OPENAI_MODEL=meta-llama/Meta-Llama-3.1-405B

Usage

Start the server:

pnpm start

This will start the server on stdio, making it available for MCP clients to communicate with.

Docker Usage

Building the Docker Image
docker build -t mcp-openai-complete .
Running the Container
# Run with environment variables
docker run -it --rm \
  -e OPENAI_API_KEY="your-api-key" \
  -e OPENAI_MODEL="gpt-3.5-turbo-instruct" \
  mcp-openai-complete

You can also use a .env file:

# Run with .env file
docker run -it --rm \
  --env-file .env \
  mcp-openai-complete
Parameters for the "complete" tool
  • prompt (string, required): The text prompt to complete
  • max_tokens (integer, optional): Maximum tokens to generate, default: 150
  • temperature (number, optional): Controls randomness (0-1), default: 0.7
  • top_p (number, optional): Controls diversity via nucleus sampling, default: 1.0
  • frequency_penalty (number, optional): Decreases repetition of token sequences, default: 0.0
  • presence_penalty (number, optional): Increases likelihood of talking about new topics, default: 0.0

Development

For development with auto-reloading:

npm run dev

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.