EngineersofAI LogoCoursesBlogResearch LabAI LettersThe LabCode BankInteractive 3DKodrEarnestJobs
Skip to main content
Interactive 3D/MCP Architecture: Model Context Protocol
Architecture Layers - 📂 Filesystem Server
HOSTstep 1
Claude Desktop
User: "Summarize README.md"
▼
MCP CLIENTstep 2
MCP Client
Parses intent → selects read_file tool
▼
MCP SERVERstep 3
Filesystem MCP Server
Receives JSON-RPC, validates path
▼
TOOL EXECUTIONstep 4
read_file("README.md")
Returns file content as string
MCP Primitives
Tools
Executable functions the LLM can call
read_file(path)
write_file(path, content)
list_directory(path)
search_files(pattern)
Resources
Data and files the LLM can read
file://local/path
directory://local/path
Prompts
Reusable prompt templates
summarize-file
analyze-code
find-todos
Server Type
MCP (Model Context Protocol) is Anthropic's open standard for connecting AI models to external tools and data sources. Hosts embed clients; clients talk to servers over JSON-RPC.

MCP Architecture: Model Context Protocol - Interactive Visualization

The Model Context Protocol (MCP) is Anthropic's open standard for connecting AI models to external tools, data sources, and services. It defines a three-layer architecture: the Host (Claude Desktop, VS Code, custom apps) embeds an MCP Client that manages sessions and speaks JSON-RPC to MCP Servers. Each server exposes three primitives: Tools (executable functions), Resources (data/files), and Prompts (reusable templates). This visualization animates a complete request flow for four server types.

  • Four server types: Filesystem, Web Search, Database, and GitHub - each with real tool and resource definitions
  • Animate a full request flow step-by-step from host to tool execution and back
  • Toggle JSON-RPC message format to see the exact wire protocol used between client and server
  • MCP Primitives panel shows tools, resources, and prompt templates for the selected server

Part of the EngineersOfAI Interactive 3D - free interactive visualizations covering every major concept in machine learning and AI engineering. Hover any element for a plain-English explanation. No code required.