Exposing Agents as MCP servers with mcp-agent
summary
Claim: agents run as MCP servers so any MCP client (Claude, Cursor) can invoke them. Implementation uses mcp-agent built on Model Context Protocol with Temporal as durable execution. Demo composes finder, writer, proofreader, fact checker, style enforcer and an orchestrator in ~100 lines of code, connects to fetch and filesystem MCP servers, launches an asynchronous workflow that produces graded_report.mmarkdown, supports polling from a client, and successfully completes with the expected graded output.
AI-written summary · about
chapters (8)
- 0:00Agents with MCP
- 1:05LSP and MCP evolution
- 3:34Agent tech stack 2025
- 8:09MCP Agent library
- 9:54Agents as MCP servers
- 12:26Async agent workflows
- 13:07Demo: MCP Agent in action
- 17:41Summary
related talks
46mMCP for LLM Agents, Infrastructure APIs & Data Services
12mMCP and the Dawn of Agentic Intelligence
36mPutting AI Agents to Work: A New Era of Open Source Connectivity with MCP
50mAgents That Talk Back: MCP, ACP, A2A, Oh my!
26mWho Let the Agent In? Securing MCP Servers in Production
24mUnderstanding the Architecture of MCP (Model Context Protocol)
30mComposable AI: Building Next-Gen AI Agents with MCP
33mThe Model Context Protocol: Connecting AI to Everything