MCP = Model Context Protocol, a framework that lets LLMs talk to external tools and sources

You should learn about RAG.

Then have whatever ai you used rewrite it with that knowledge.

So obviously written by AI. E.g.:

"Documentation is where MCP quietly does its best work, and where much of the criticism misses the mark."

Slop. One of the arguments is that an MCP can do prompt injection via the tool descriptions. How the fuck is this on the front page.

When I see anything that says "using it wrong" and saw the number of security issues and exploited MCP servers, that was everything I needed to know about how terrible MCP is as a standard.

Just as bad, if not worse than JWTs.

JWTs are fine.

Just a note for everyone on HN. I am now hiding all LLM related stories like this one and have no regrets. Actually my front page looks pretty good now, with varied and interesting stories like back in the day.

so yes, it was clearly drafted by Claude. But I stand by the ideas, which I gave Claude to write this. We have a mix of automated and manual articles on the site, and the automated ones are all labeled Claude, and some people have been enjoying them. Happy to remove if against HN guidelines to have any claude-text.