# Model Context Protocol (MCP)

An open protocol that gives an AI application a standard way to discover and call external tools and data sources, using a host, client and server arrangement.

Source: https://aecon.ai/glossary/model-context-protocol
Updated: 2026-09-15

## In practice

A chat application connects to an MCP server for a company's document store, lists the tools it offers, and calls one to retrieve a policy the model then cites.

## The distinction

MCP standardises the conversation, not trust. A server can be perfectly protocol-conformant and still be unsuitable for a sensitive job, wrongly scoped, unavailable or operated by someone the buyer has never checked.

## Three roles

A **host** is the AI application the person is using. A **client** is the connector inside that host. A **server** exposes resources, prompts and tools that the client can list and invoke.

The separation matters because the model never reaches the outside world directly. It proposes a call; the application executes it. That boundary is where argument validation, permission checks and approval belong.

## What it standardises, and what it does not

| MCP supplies | The buyer still supplies |
| --- | --- |
| A common way to list and invoke tools | Whether this server is trusted for this job |
| Input schemas and structured results | Server-side validation of what the model proposed |
| Capability negotiation and versioning | A pinned server identity and version |
| Optional OAuth-based authorisation | The policy deciding whether the action is permitted |

The specification is explicit that consent and human control sit with the host and client, and that the protocol cannot enforce those principles at the protocol level. A registry listing or a familiar tool name is discovery evidence, not an approval.

## Reading it in the market

MCP has become a common way to describe tools, which is why connector catalogues and registries grew quickly around it. Those counts measure published supply. They do not measure how often independent customers invoke a server, whether the work was accepted, or whether the route is safe in a particular deployment.

## Sources & context

- [Model Context Protocol specification](https://modelcontextprotocol.io/specification/2026-07-28/) — Model Context Protocol, 2026-07-28. Defines host, client and server roles, tool discovery and invocation over JSON-RPC, and states that tool annotations should be treated as untrusted unless the server is trusted. Checked 2026-09-13.
- [MCP Registry](https://modelcontextprotocol.io/registry/about) — Model Context Protocol, Preview, 2026. A preview metadata service that manages namespaces and points to separately published packages; it does not host artifacts or assess security and uptime. Checked 2026-09-13.

## Related reading

- [Tool](https://aecon.ai/glossary#tool)
- [AI agent](https://aecon.ai/glossary/ai-agent)
- [Agent harness](https://aecon.ai/glossary/agent-harness)
- [Agent2Agent (A2A)](https://aecon.ai/glossary/agent2agent-protocol)