MCP resources and prompts
MCP resources and prompts are two things an MCP server can offer besides tools. Resources are pieces of data, each with a URI, that an app can add to context; prompts are reusable message templates a user chooses to run.
Also called: MCP primitives
By Divya Sharma · Checked Nov 15, 2026 · How we review
Most people meet MCP through tools, which the model calls to take actions. The protocol also defines two other server features, and together the three are often called MCP primitives. They differ mainly in who decides when they are used.
Resources
A resource is data the server makes readable: a file, a database schema, a log, a record in an app. Each one has a URI such as file:///project/src/main.rs and a MIME type. The MCP client can list resources, read them, and subscribe to be told when one changes. The specification calls resources application-driven: the host app decides how they enter the conversation, for example through a picker where the user attaches a file, or by adding them automatically.
Prompts
A prompt is a template the server publishes with a name, a description and optional arguments. When a user picks it, the client fetches the filled-in messages and sends them to the model. Prompts are user-controlled, and apps often show them as slash commands. A code review server might offer a review prompt that takes the code as its argument.
How they compare with tools
Tools are invoked by the model during function calling. Resources bring in context without an action. Prompts package a task the user starts. An MCP server can offer any mix of the three.
Limits
Client support is uneven. Many apps use only tools, so a server's resources and prompts may never appear. Check what your client supports in the MCP tools directory reviews.
Related terms
- MCP server
- An MCP server is a program that exposes tools, data resources or prompt templates to AI applications through the Model Context Protocol, so any MCP-compatible client can use them without custom integration code.
- Tool calling
- Tool calling, also called function calling, is a model API feature where you describe functions with a schema, the model replies with a structured request to call one, and your code runs it and returns the result.
- MCP client
- An MCP client is the part of an AI application that holds one connection to one MCP server, negotiating capabilities and carrying requests, results and notifications between that server and the host app.
- Model Context Protocol
- The Model Context Protocol (MCP) is an open standard that defines how AI applications connect to external tools, data and prompts, using JSON-RPC messages between a host application, its clients and separate servers.
- Slash command
- A slash command is an instruction typed with a leading slash, such as /compact or /review, that tells an AI tool to run a built-in action or a saved prompt instead of treating the text as an ordinary message.