Top news
Investing in the public benefit OfficialAnthropic NewsOct 18Popular Musk Pledges SpaceX Shares to ‘Every Kid in America’ WebBloomberg TechnologyOct 19Popular Websites are closing their doors to AI agents WebThe Verge AIOct 17 This is the new ChatGPT Voice, powered by GPT Live OfficialOpenAI on YouTubeOct 18 How AI giants will reshape global philanthropy WebAxios AIOct 19 Gemini 4 Argon Really Surprised Me WebTwo Minute PapersOct 19 Google Gemini 4 agents just beat Minecraft with no human intervention WebEngadget AIOct 19 Who could it be? 🧐 OfficialGoogle on YouTubeOct 18 Anthropic gives outside AI watchdog “employee-like” access WebArs Technica AIOct 18 What are self-replicating ‘AI worms’, and why is Anthropic worried about them? WebThe Guardian AIOct 18
GlossaryProductsIntermediate

Model context protocol

Also written: MCP, MCP (Model Context Protocol), Bidirectional MCP, MCP configurations, MCP integration, Microsoft Copilot integration, MCP Events specification, MCP Events

A protocol enabling AI models to connect to external services and tools, allowing them to read from and write to them.

In plain terms

Model Context Protocol (MCP) is a way for AI models to connect to and use other applications and services. Think of it like giving your AI a way to "look up" information or "take action" in other programs, just like you might.

For example, you could use MCP to let an AI read your calendar to schedule a meeting, or write notes into a document. This helps the AI do more complex tasks by working with the tools you already use.

The expert version

MCP is a protocol that defines how AI models interact with external services and tools, enabling them to access and manipulate data beyond their internal knowledge. It facilitates the integration of AI capabilities with third-party applications, databases, and APIs, effectively extending the model's operational scope. This allows for more dynamic and context-aware AI behavior, where models can retrieve real-time information or execute actions based on external data.

Why it matters now

Model Context Protocol (MCP) is gaining traction as AI models evolve beyond simple text generation to actively interact with external tools and data. Recent announcements highlight MCP's role in enabling AI agents to run within private clusters, integrate with business systems, and enhance code development workflows. This shift reflects a broader trend towards more capable and integrated AI assistants.

Where it shows up 51 items, newest first
How each company uses it
Google Gemini
Google is integrating MCP into Apigee hybrid for agentic tools in private clusters and is bringing agentic AI to Gemini, enabling it to act as a coworker and handle tasks.
Microsoft Copilot
Microsoft is introducing the Agent Experience (AX) Practitioner Playbook, which likely details how AI coding agents, potentially using MCP, can better interact with their technology.
OpenAI ChatGPT
OpenAI's ChatGPT plugins, like the one that dramatically boosted adoption for Stack Overflow's agent product, demonstrate a practical application of models interacting with external tools and data, aligning with MCP principles.
xAI Grok
xAI Grok is enabling bots and humans to fix code together in one thread, suggesting an integration where AI agents can interact with development tools and environments, likely facilitated by protocols like MCP.
Related terms

AgentsTool useAPIsContext windowFunction callingInteroperability

Often confused with
Context window
The context window refers to the amount of information an AI can
APIs
APIs are interfaces that allow software to communicate, while MCP is a spe
Plugins
Plugins are extensions that add functionality to an AI, often using pro
Function calling
Function calling is a technique where a model can output a req
Origin

The concept of protocols for AI to interact with external systems has evolved with the rise of AI agents and tool use. Specific implementations and naming conventions are emerging from frontier AI labs as they build more integrated AI products.

Try this about 5 minutes

Connect an app to Google Gemini by going to its settings and exploring the 'Connected apps' section. You can then ask Gemini to perform tasks that involve that app, like summarizing information from it or scheduling an event within it. This demonstrates how Gemini uses a protocol similar to MCP to interact with external services.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

What is the main benefit of MCP?
MCP allows AI models to access and use information from external applications and services, making them more capable and versatile.
How does MCP help AI remember things?
MCP doesn't directly help AI remember information within its own memory. Instead, it allows the AI to retrieve or write information to external systems, which can then be used as context.
Is MCP the same as an API?
No, MCP is a protocol that defines how AI models should use APIs and other services to interact with their environment. APIs are the underlying communication interfaces.

First seen on Subvolts Apr 3, 2025. Updated Oct 16, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.