Cryptographer Criticizes Local MCP Server Design
matthew_d_green · x · 2026-07-20
Matthew Green believes that while AI agent designers praise running local MCP servers to give models native skill sets, this design lacks actual testing (dogfooding).
He pointed out that agents themselves do not want these clunky local servers. More importantly, users often have to restart the session for the agent to invoke the MCP server, pushing the tool-calling process straight into a usability wall and severely degrading the experience.
More from coding & agent
- Tenable and AWS launch a Black Hat build event for open-source security agents and MCP servers — Dave_Maynor · 2026-07-22
- Codex helps build Valdiluce, an open-world game with climbing, gliding and gondolas — Dimillian · 2026-07-22
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22
- Agent search bottlenecks are now about variance, not raw latency — rohanpaul_ai · 2026-07-22
- LangSmith adds tracing for Pipecat, LiveKit, OpenAI Realtime, and Gemini Live — LangChain · 2026-07-22
- An MCP server signs every AI agent tool call into a verifiable Merkle chain — Funky_Chicken_22 · 2026-07-22