Letting AI Interface with Your App with MCPs
summary
Claims how to expose app functionality via MCPs and shows a working demo: an MCP server network calls 17 tools, a local MCP server for GPS, a weather MCP on Cloudflare using Acueeather API, and an EpicMe MCP for journal entry creation. The LLM chains tool outputs, prompts OAuth email validation token for auth, writes and tags a journal entry, returns JSON rendered as markdown, and successfully deletes and logs out — proving the end-to-end flow works today.
AI-written summary · about
chapters (6)
- 0:00AI assistants and MCPs
- 3:11Current AI integration limits
- 8:07Model Context Protocol
- 11:37Live MCP demo
- 16:53MCP benefits and future
- 18:15Resources and closing
related talks
37mLetting AI Interface with Your App with MCPs
20mLetting AI Interface with your App with MCP
35mMCP in Action: Real-World Use Cases & Design Patterns
25mMaking AI Personal: Building Context-Rich AI Tools with MCP
26mBuild your First MCP App
29mMCP Apps: Primitives, discovery, and the Future of Software
16mOne AI Chat to Rule Them All: Connecting Enterprise Tools with MCP
47mMCP Demystified: Cutting Through the Hype-What It Is, What Isn't & Its Role in the Agentic RAG Stack