MCP was always a bad idea?
Simon Willison published a comment responding to a Hacker News article titled "MCP was always a bad idea?", saying the article entirely misses the value that MCP brings today.
Willison wrote that there is almost no reason to use MCPs if you are running a full-blown terminal agent with unfettered internet access, naming Claude Code, Codex, Meta Muse and OpenClaw as examples, and said in that case you can just let the agent call APIs directly.
He said that if you want to operate something less "YOLO" than that, you will find yourself wanting four things: control over exactly which external services the agent can access; a way to handle authentication that does not allow the agent to directly access API keys; a sensible UI to allow users to connect and authenticate further services; and strong audit logging for what is going on.
According to Willison, MCP makes all of that so much easier to provide. He concluded that thinking MCP is obsolete because full coding agents do not need it misses out on all of the other things we might want to build.
The post is tagged hacker-news and model-context-protocol.
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
My comment on MCP was always a bad idea? — Hacker News. This article entirely misses the value that MCP brings today. Sure, there's almost no reason to use MCPs if you are running a full-blown terminal agent (Claude Code, Codex, Meta Muse, OpenClaw etc) with unfettered internet access - just let it call APIs directly. If you want to operate something that's less YOLO than that, you'll find yourself wanting: Control over exactly which external services it can access A way to handle authentication that doesn't