How to build MCP Clients | MCP Trilogy | CampusX

Description

In this video, we learn how to build your own MCP client and connect it to multiple MCP servers — including a local Math server, a remote Expense Tracker server, and even a Manim animation server. We go step-by-step through the setup, tool-calls, and finally create a clean Streamlit-based chatbot UI that can talk to all these servers. Perfect for anyone learning MCP, LangChain, or custom AI tool integrations. Code: https://github.com/campusx-official/mcp-client Quick Revision: https://docs.google.com/document/d/1I26xHGosv0B7zgDl0sM_GZtlX20rPWYBPZHEcoREN7A/edit?usp=sharing Learn LangGraph: https://www.youtube.com/playlist?list=PLKnIA16_RmvYsvB8qkUQuJmJNuiCUJFPL Config Part: https://youtu.be/y-uPv3ltOTY?t=915&si=ObvzsrnWJZjDkG33 Async/Await Tutorial: https://realpython.com/async-io-python/ ============================ Did you like my teaching style? Check my affordable mentorship program at : https://learnwith.campusx.in DSMP FAQ: https://docs.google.com/document/d/1OsMe9jGHoZS67FH8TdIzcUaDWuu5RAbCbBKk2cNq6Dk/edit#heading=h.gvv0r2jo3vjw ============================ 📱 Grow with us: CampusX' LinkedIn: https://www.linkedin.com/company/campusx-official CampusX on Instagram for daily tips: https://www.instagram.com/campusx.official My LinkedIn: https://www.linkedin.com/in/nitish-singh-03412789 Discord: https://discord.gg/PsWu8R87Z8 E-mail us at support@campusx.in ⌚Chapters⌚ 00:00 — Intro & playlist recap (apology + what we covered so far) 00:28 — MCP playlist structure & approach (WHY / WHAT / HOW) 03:54 — Video goal & servers overview (what we’ll build — Math + Expense servers) 06:13 — Local Math MCP server demo & testing (uvicorn, inspector) 08:07 — Project setup & dependencies (folder, uv init, install libs) 09:23 — Build client: skeleton, async structure & server config 14:06 — Tools & LLM binding (fetch tools, name-tools dict, bind OpenAI LLM) 19:18 — Tool-call flow & invocation (extract tool calls, invoke tools, get results) 23:32 — Returning tool results to LLM & final response (ToolMessage + history) 29:29 — Multi-server support (loop handling + add remote Expense Tracker) 33:37 — External servers & Manim animation integration (generate animations) 35:21 — Streamlit GUI: convert logic to UI and live demo 39:25 — Summary, next steps (LangGraph playlist) & outro