OpenAI Assistant MCP

Example Use Cases

• 1

  Create, list, and update OpenAI Assistants programmatically via the Model Context Protocol.

• 2

  Maintain persistent, named conversation threads across sessions with local SQLite-backed metadata.

• 3

  Streaming, low-latency assistant responses to display real-time progress and avoid request timeouts.

Description

This server exposes tools to create, list, retrieve, and update OpenAI Assistants and to start and manage persistent conversation threads. It stores thread metadata locally in SQLite so threads can be named, listed, reused, and deleted, compensating for OpenAI's lack of thread listing. The primary conversation tool streams assistant responses in real time to avoid timeouts and provide immediate progress updates. It is designed to integrate with clients like Claude Desktop and requires an OpenAI API key for operation.