Active Standard Version 1.0

Model Context Protocol (MCP)

Anthropic's standard for connecting AI models to data sources and tools, used in the ecosystem for tool integration and context management.

MCPJSON-RPC

Overview

The Model Context Protocol (MCP) is an open standard that enables AI models to securely interact with external data and tools. In the Pragma.Vision ecosystem, it serves as the foundational layer for tool exposure and context retrieval.

Core Concepts

Clients, Hosts, and Servers

  • Hosts: Applications like Claude Desktop or IDEs that initiate connections.
  • Clients: Maintain 1:1 connections with servers.
  • Servers: Provide context, prompts, and tools to the client.

Resources, Prompts, and Tools

  • Resources: File-like data (logs, code, database records) that can be read by clients.
  • Prompts: Pre-written templates that help users accomplish specific tasks.
  • Tools: Executable functions that can perform actions (e.g., querying a database).

Integration in Pragma.Vision

We use MCP to expose our ecosystem’s capabilities to AI agents:

  • Research Access: Indexing research papers as MCP resources.
  • Standard Verification: Tools to verify compliance with AP2/ACP/TAP.
  • Marketplace Data: Reading agent listings via MCP.

Security Features

  • Controlled Access: Users must explicitly approve tool execution and resource access.
  • Transport Agnostic: Works over Stdio, HTTP/SSE.

Resources