Glossary

Model Context Protocol (MCP): what it is and why it matters

The Model Context Protocol (MCP) is an open standard for connecting AI applications to external tools, data and systems through one common interface. Anthropic introduced it in November 2024, and it is now supported across major AI clients and platforms. For a business, MCP means a system connected once can be used by any compatible model or agent.

Shaun DevanFeaturing insights fromShaun Devan · Founder & CEO

Updated · 3 min read

Get in touchSee our work

MCP vs a custom integration

Custom integrationMCP server
Built forOne model or one appAny MCP-compatible client
Changing modelsRewrite the integrationKeep the server, swap the client
What it exposesWhatever the code doesDefined tools, resources and prompts
Access controlAd hocSet at the server, per tool

How it works

An MCP server wraps a system, such as a CRM, a database or a file store, and exposes what an AI may do with it as named tools and readable resources. An MCP client, such as an AI assistant or an agent, discovers those tools and calls them. The server decides what is allowed, so read-only access, approval steps and logging live in one place.

Why businesses care

  • Connect a system once and reuse it across agents and assistants.
  • Switch models without rebuilding integrations.
  • Keep permissions and audit trails at the system boundary, where your IT team can see them.

Sources

Frequently asked questions

Who created the Model Context Protocol?

Anthropic released MCP as an open standard in November 2024. It is now an open project with a public specification and SDKs in several languages.

Is MCP the same as an API?

No. An API is how any software talks to a system. MCP is a standard way for AI applications to discover and use tools, and an MCP server usually calls the system's API underneath.

Is MCP secure?

MCP itself is a protocol; security depends on how servers are built and deployed. Good practice is least-privilege tools, read-only by default, approval for writes, and logging of every call.

Keep reading

InsightThe Model Context Protocol, explained for business and IT leadersMCP is an open standard for connecting AI agents to the systems a business runs. What it does, when it helps, the security questions, and how to start.GlossaryAI agent: what it is, and how it differs from a chatbotAn AI agent is software that uses a language model to pursue a goal: it plans steps, calls tools and systems, and acts on the results.GlossaryContext engineering: what it is and why it replaced prompt tuningContext engineering is designing everything a model sees for a task: instructions, examples, data, tool results and memory, so an agent acts reliably.GlossaryAI orchestration: how multiple agents work togetherAI orchestration coordinates AI agents, models and tools so they run as one system: who does each step, in what order, and what happens on failure.NativeSecurity

Put AI to work across your whole business.

Bring the hardest problem on your list. In 30 minutes we'll show you where AI will have the most impact first, and what it takes to get there.

Book a callSee our work