---
Title: "What Is Model Context Protocol (MCP)? Explained"
Url: "https://computer.io/glossary/model-context-protocol"
Topic: "Agentic AI"
Description: "Model Context Protocol (MCP) is an open standard for connecting AI models and agents to tools and data. How MCP works, examples and MCP vs APIs."
---

# What is Model Context Protocol (MCP)?

Model Context Protocol (MCP) is an open standard, introduced by Anthropic in 2024, for connecting AI models and agents to external tools and data sources. MCP servers expose tools and resources in a common format, so any MCP-compatible AI application can use them without custom integrations.

## How it works

| Part | Role |
| --- | --- |
| MCP host | AI app (assistant, IDE, agent) |
| MCP client | Connects to servers |
| MCP server | Exposes tools, resources, prompts |

## Example

An agent connects to a ticketing MCP server and calls "search_tickets" and "update_ticket" tools.

## MCP vs API

APIs are system-specific. MCP is a standard layer that lets AI discover and call tools consistently.

## Considerations

Authentication, permissions, and limiting what tools an agent can call.

## FAQ

### Who created MCP?

Anthropic introduced it as an open standard in November 2024.

### Is MCP only for Claude?

No, many AI tools and platforms support it.

### Is MCP secure?

It depends on implementation; use auth, scopes and approvals.

## Related terms

- [Tool calling (function calling)](https://computer.io/glossary/tool-calling)
- [Agent2Agent protocol (A2A)](https://computer.io/glossary/agent2agent-protocol)
- [AI agent](https://computer.io/glossary/ai-agent)
- [How does MCP connect AI agents to business tools?](https://computer.io/answers/how-does-mcp-connect-ai-agents-to-business-tools)

## How Computer helps

- [How computer works](https://computer.io/how-computer-works)

## Compare

- [Devrev vs claude](https://devrev.ai/compare/devrev-vs-claude)

## Further reading

- [Knowledge graph vs mcp interfaces](https://devrev.ai/blog/knowledge-graph-vs-mcp-interfaces)