Build AI-Powered Automation Tools: MCP Workshop for Developers & Engineers
summary
Workshop teaches MCP fundamentals and walks through a TypeScript MCP server that exposes add and multiply tools plus a resource, communicates via JSON‑RPC over STDIO or HTTP, and integrates with clients like Cursor/ru using Gemini or local LM Studio models. Demo shows the server running after fixing output schema, an addition call costing ≈$0.02 and using ~27.8k tokens of a 1.0M window. Covers real risks — prompt injection, tool poisoning, tool shadowing — and a mitigation SecureHulk interceptor that scans and intercepts tool calls.
AI-written summary · about
chapters (8)
- 0:00MCP introduction and overview
- 4:48Need for MCP servers
- 13:23MCP architecture explained
- 33:21TypeScript MCP server setup
- 54:44Building MCP tools
- 1:03:16MCP server security risks
- 1:21:35Securing MCP servers
- 1:31:00Q&A
related talks
25mMaking AI Personal: Building Context-Rich AI Tools with MCP
01h 01m.NET Meets MCP: Build Your Own AI-Powered Service with C#
25mTools in your AI's Toolbox : An introduction to MCP Servers
35mMCP in Action: Real-World Use Cases & Design Patterns
11mBuilding MCP Apps w/ All the Comfort of Modern TypeScript DevX
52mBeyond local tools: Deep dive into MCP with Spring AI
22mNext-Gen AI Tooling with MCP Servers in Go
17mBuilding Production-Ready MCP Servers: From AI Agents to Secure Enterprise APIs.