Notes
Claude
Introduction to MCP
Connecting with MCP Clients
Accessing Resources

Accessing Resources

Resources in MCP allow your server to expose information that can be directly included in prompts, rather than requiring tool calls to access data. This creates a more efficient way to provide context to AI models.

Example

Sequence diagram of resource access: a user types “What’s in the @…” in the CLI, our code sends a ReadResourceRequest to the MCP server, and the server returns a ReadResourceResult containing the resource content

When a user types something like “What’s in the @…” our code recognizes this as a resource request, sends a ReadResourceRequest to the MCP server, and gets back a ReadResourceResult with the actual content.

Implementing Resource Reading

To enable resource access in your MCP client, you need to implement a read_resource function.

import json
from pydantic import AnyUrl

async def read_resource(self, uri: str) -> Any:
    # Make a request to the MCP server and processes the response based on its MIME type
    result = await self.session().read_resource(AnyUrl(uri))
    resource = result.contents[0]

    if isinstance(resource, types.TextResourceContents):
        if resource.mimeType == "application/json":
            return json.loads(resource.text)

    return resource.text

Understanding the response structure

When you request a resource, the server returns a result with a contents list.

We access the first element since we typically only need one resource at a time. The response includes:

  • The actual content (text or data)
  • A MIME type that tells us how to parse the content
  • Other metadata about the resource

Content type handling

The function checks the MIME type to determine how to process the content:

  • If it’s application/json, parse the text as JSON and return the parsed object
  • Otherwise, return the raw text content

This approach handles both structured data (like JSON) and plain text documents seamlessly.

docs