README.md

layout	default
title	MCP Use Tutorial
nav_order	163
has_children	true
format_version	v2

MCP Use Tutorial: Full-Stack MCP Development Across Agents, Clients, Servers, and Inspector

Learn how mcp-use/mcp-use composes agent, client, server, and inspector workflows across Python and TypeScript with practical security and operations patterns.

Why This Track Matters

mcp-use is one of the most comprehensive open-source MCP ecosystems, covering agent orchestration, client transport handling, server frameworks, and web-based inspection in one stack.

This track focuses on:

• selecting Python or TypeScript workflows based on team context
• configuring multi-server clients and agent guardrails
• building server-side capabilities with inspector-first feedback loops
• hardening security and production behavior for real deployments

Current Snapshot (auto-updated)

• repository: mcp-use/mcp-use
• stars: about 9.7k
• latest release: python-v1.7.0 (published 2026-03-17)

Mental Model

flowchart LR
    A[LLM] --> B[MCPAgent]
    B --> C[MCPClient]
    C --> D[MCP servers]
    D --> E[Tools resources prompts]
    C --> F[Inspector workflows]

Loading

Chapter Guide

Chapter	Key Question	Outcome
01 - Getting Started and Stack Selection	Should I start with Python or TypeScript and how fast can I ship?	Faster onboarding
02 - Client Configuration, Sessions, and Transport Choices	How do I configure robust client connectivity across servers?	Reliable client baseline
03 - Agent Configuration, Tool Governance, and Memory	How do I keep agent behavior useful and controlled?	Better agent discipline
04 - TypeScript Server Framework and UI Widgets	How do I build rich MCP servers with TypeScript + UI resources?	Strong TS server workflow
05 - Python Server Framework and Debug Endpoints	How do I build and inspect MCP servers in Python?	Strong Python server workflow
06 - Inspector Debugging and Chat App Workflows	How does inspector improve developer and QA loops?	Faster issue isolation
07 - Security, Runtime Controls, and Production Hardening	How do I reduce risk when agents can access powerful tools?	Safer production posture
08 - Operations, Observability, and Contribution Model	How do teams sustain and evolve mcp-use usage?	Long-term execution plan

What You Will Learn

• how to use mcp-use as a full-stack MCP platform, not just a thin client
• how to configure multi-server, multi-language workflows with clearer boundaries
• how to apply security and access control patterns to high-power tool chains
• how to run contribution and operational loops that keep behavior predictable

Source References

• mcp-use Main README
• TypeScript README
• Python README
• TypeScript Quickstart
• Python Quickstart
• TypeScript Client Config
• TypeScript Server Config
• Python Server Intro
• Inspector Docs

Related Tutorials

• MCP TypeScript SDK Tutorial
• MCP Python SDK Tutorial
• MCP Inspector Tutorial
• FastMCP Tutorial

---

Start with Chapter 1: Getting Started and Stack Selection.

Navigation & Backlinks

• Start Here: Chapter 1: Getting Started and Stack Selection
• Back to Main Catalog
• Browse A-Z Tutorial Directory
• Search by Intent
• Explore Category Hubs

Full Chapter Map

1. Chapter 1: Getting Started and Stack Selection
2. Chapter 2: Client Configuration, Sessions, and Transport Choices
3. Chapter 3: Agent Configuration, Tool Governance, and Memory
4. Chapter 4: TypeScript Server Framework and UI Widgets
5. Chapter 5: Python Server Framework and Debug Endpoints
6. Chapter 6: Inspector Debugging and Chat App Workflows
7. Chapter 7: Security, Runtime Controls, and Production Hardening
8. Chapter 8: Operations, Observability, and Contribution Model

Generated by AI Codebase Knowledge Builder