Model Context Protocol is an evolving standard that allows LLMs to connect to external data sources and tools.
If you're familiar with
The Model Context Protocol (MCP) is an open standard developed by Anthropic in November 2024 to facilitate seamless integration between AI applications and external data sources and tools.
It provides a standardized interface that allows AI models to access and interact with various resources, enhancing their capabilities and contextual understanding. Philschmid+9Medium+9Wikipedia, l'enciclopedia libera+9arXiv
OpenAI has recognized the value of MCP and is actively supporting its integration across its products.
In March 2025, OpenAI CEO Sam Altman announced plans to incorporate MCP support into OpenAI's offerings, including the ChatGPT desktop app and the Agents SDK.
This integration aims to provide users with a more unified and efficient experience when connecting AI models to external data sources. PromptHub+2TechCrunch+2Medium+2
The Agents SDK by OpenAI has been updated to include support for MCP, enabling developers to utilize a wide range of MCP servers to provide tools to their AI agents. This allows for more dynamic and context-aware interactions within AI applications. GitHub+5OpenAI GitHub+5arXiv+5
By adopting MCP, OpenAI aims to enhance the interoperability of its AI models with external systems, promoting a more standardized and efficient approach to integrating AI assistants with various data sources and tools.OpenAI GitHub+4Medium+4Medium+4
Differences from Tools
Before MCP, OpenAI used its own proprietary interface for tool use, often referred to as function calling or tool calling, especially in the ChatGPT and GPT API ecosystem.
FAQs
Answers to frequently asked questions.
