MCP (Model Context Protocol)
An open standard for connecting AI agents to outside tools and data sources, such as databases, APIs or documents, through one kind of connector.
Without MCP, every agent would need its own integration with every tool. With MCP, an MCP server exposes a source (your database, say) and any compatible agent can use it.
What matters in data work: an MCP connector gives the agent real access. Before you install one, answer in writing:
- What can it read? The whole database, or one schema?
- What can it change? If you only need queries, make the credential read-only.
- Which credential does it use, and who else uses it?
- Where do the results go? What the connector returns enters the model's context.
Common confusion: "the agent will only do what I ask". The connector doesn't know what you asked; it knows what the credential allows. Never connect an agent to production "to try it out".
Learn more
Related terms
- AI agent
A program that pairs a language model with tools, such as terminal commands and files, to carry out a task in several steps: it reads its context, acts, looks at the result and repeats.
- Agent harness
The program that wraps a language model and makes it an agent: it gives the model tools, asks for your permission, keeps the session and decides what goes into its context.
- Context window
Everything a language model has in front of it at a given moment: your instruction, the files it has read and the results of its tools. What isn't in the context, the model can't use.