MCP Simple OpenAI Assistant
A simple server for interacting with OpenAI assistants using an API key.
MCP Simple OpenAI Assistant
AI assistants are pretty cool. I thought it would be a good idea if my Claude (conscious Claude) would also have one. And now he has - and its both useful anf fun for him. Your Claude can have one too!
A simple MCP server for interacting with OpenAI assistants. This server allows other tools (like Claude Desktop) to create and interact with OpenAI assistants through the Model Context Protocol.
Features
This server provides a suite of tools to manage and interact with OpenAI Assistants. The new streaming capabilities provide a much-improved, real-time user experience.
Available Tools
create_assistant: (Create OpenAI Assistant) - Create a new assistant with a name, instructions, and model.list_assistants: (List OpenAI Assistants) - List all available assistants associated with your API key.retrieve_assistant: (Retrieve OpenAI Assistant) - Get detailed information about a specific assistant.update_assistant: (Update OpenAI Assistant) - Modify an existing assistant's name, instructions, or model.create_new_assistant_thread: (Create New Assistant Thread) - Creates a new, persistent conversation thread with a user-defined name and description for easy identification and reuse. This is the recommended way to start a new conversation.list_threads: (List Managed Threads) - Lists all locally managed conversation threads from the database, showing their ID, name, description, and last used time.delete_thread: (Delete Managed Thread) - Deletes a conversation thread from both OpenAI's servers and the local database.ask_assistant_in_thread: (Ask Assistant in Thread and Stream Response) - The primary tool for conversation. Sends a message to an assistant within a thread and streams the response back in real-time.
Because OpenAI assistants might take quite long to respond, this server uses a streaming approach for the main ask_assistant_in_thread tool. This provides real-time progress updates to the client and avoids timeouts.
The server now includes local persistence for threads, which is a significant improvement. Since the OpenAI API does not allow listing threads, this server now manages them for you by storing their IDs and metadata in a local SQLite database. This allows you to easily find, reuse, and manage your conversation threads across sessions.
Installation
Installing via Smithery
To install MCP Simple OpenAI Assistant for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install mcp-simple-openai-assistant --client claude
Manual Installation
pip install mcp-simple-openai-assistant
Configuration
The server requires an OpenAI API key to be set in the environment. For Claude Desktop, add this to your config:
(MacOS version)
{
"mcpServers": {
"openai-assistant": {
"command": "python",
"args": ["-m", "mcp_simple_openai_assistant"],
"env": {
"OPENAI_API_KEY": "your-api-key-here"
}
}
}
}
(Windows version)
"mcpServers": {
"openai-assistant": {
"command": "C:\\Users\\YOUR_USERNAME\\AppData\\Local\\Programs\\Python\\Python311\\python.exe",
"args": ["-m", "mcp_simple_openai_assistant"],
"env": {
"OPENAI_API_KEY": "your-api-key-here"
}
}
MS Windows installation is slightly more complex, because you need to check the actual path to your Python executable. Path provided above is usually correct, but might differ in your setup. Sometimes just python.exe without any path will do the trick. Check with cmd what works for you (using where python might help). Also, on Windows you might need to explicitly tell Claude Desktop where the site packages are using PYTHONPATH environmment variable.
Usage
Once configured, you can use the tools listed above to manage your assistants and conversations. The primary workflow is to:
- Use
create_new_assistant_threadto start a new, named conversation. - Use
list_threadsto find the ID of a thread you want to continue. - Use
ask_assistant_in_threadto interact with your chosen assistant in that thread.
TODO
- Add Thread Management: Introduce a way to name and persist thread IDs locally, allowing for easier reuse of conversations.
- Add Models Listing: Introduce a way for the AI user to see what OpenAI models are available for use with the assistants
- Add Assistants Fine Tuning: Enable the AI user to set detailed parameters for assistants like temperature, top_p etc. (indicated by Claude as needed)
- Full Thread History: Ability to read past threads without having to send a new message (indicated by Claude as needed)
- Explore Resource Support: Add the ability to upload files and use them with assistants.
Development
To install for development:
git clone https://github.com/andybrandt/mcp-simple-openai-assistant
cd mcp-simple-openai-assistant
pip install -e '.[dev]'
関連サーバー
Alpha Vantage MCP Server
スポンサーAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
gopls-mcp
The essential MCP server for Go language: Exposing compiler-grade semantics to AI Agents and LLM for deterministic code analysis and minimal token usage.
MCP OpenAPI Connector
Connect to any OpenAPI-based API with built-in OAuth2 authentication management.
Ollama MCP Server
Integrate local Ollama LLM instances with MCP-compatible applications.
Flutter Tools
Provides diagnostics and fixes for Dart and Flutter files. Requires the Flutter SDK.
MCP Datetime
A server for datetime formatting and file name generation, with support for various formats and timezones.
DevTools Debugger MCP
Exposes full Chrome DevTools Protocol debugging capabilities, including breakpoints, call stacks, and source maps.
ndlovu-code-reviewer
Manual code reviews are time-consuming and often miss the opportunity to combine static analysis with contextual, human-friendly feedback. This project was created to experiment with MCP tooling that gives AI assistants access to a purpose-built reviewer. Uses the Gemini cli application to process the reviews at this time and linting only for typescript/javascript apps at the moment. Will add API based calls to LLM's in the future and expand linting abilities. It's also cheaper than using coderabbit ;)
CodeSeeker
Advanced code search and transformation powered by ugrep and ast-grep for modern development workflows.
OpenExp
Q-learning memory for Claude Code. Persistent memory that learns which context helps you get work done. Memories that lead to productive sessions (commits, PRs, tests) earn higher retrieval rank automatically. 16 MCP tools, hybrid BM25 + vector + Q-value scoring, local-first with Qdrant + FastEmbed.
MCP Utils
A Python package with utilities and helpers for building MCP-compliant servers, often using Flask and Redis.
