MCP clients

Purpose and Role

• Client provides communication between your server and an MCP server
• Acts as the access point to all tools implemented by the MCP server

Transport Agnostic Communication

• MCP can communicate over multiple protocols (not locked to one method)
• Common setup: client and server on same machine communicating via standard input/output
• Alternative protocols: HTTP, web sockets, and others

Message Types

• List Tools Request: sent from client to server; asks server to list all available tools
• List Tools Result: server response containing list of all tools it provides
• Call Tool Request: asks server to run a specific tool with particular arguments
• Call Tool Result: contains the result of the tool execution

Typical Communication Flow

• User submits query to server
• Server asks MCP client for list of tools
• MCP client sends list tools request to MCP server
• MCP server responds with list tools result
• Server sends user query + tool list to Claude (or other LLM)
• Claude decides to use a tool and responds with tool use message
• Server asks MCP client to execute the tool
• MCP client sends call tool request to MCP server
• MCP server executes tool (e. g. , calls GitHub API)
• MCP server wraps result in call tool result and sends back
• MCP client returns result to server
• Server sends tool result to Claude for final response generation
• Claude formulates answer and sends back to user

Takeaways

• MCP client acts as the communication bridge between your server and an MCP server, providing access to all available tools
• MCP supports multiple transport protocols (stdin/stdout, HTTP, web sockets) rather than being locked to one method
• Communication involves four key message types: List Tools Request/Result and Call Tool Request/Result
• The typical flow involves the server requesting available tools from the MCP server, passing them to Claude, executing Claude's chosen tool via the MCP client, and returning results for final response generation