Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Stateless MCP has recaptured my interest

simonwillison.net

Post reply on HN
Print view View original

141–150 of 247 posts

Re: Stateless MCP has recaptured my interest

#141

Post by wolttam »

lexicality wrote: post #105
ai_critic wrote: post #50

It is amazing that folks getting paid several hundred thousand dollars a year...reinvented RPC-over-HTTP/JSON. You too, fellow web developers, are smart enough to work at Anthropic. I would love to see a proper engineering post-mortem for how this happened.

They didn't reinvent it, it's literally using JSONRPC. The thing they "invented" was coming up with a loose standard for how to implement your RPC endpoints in a way that the AI was pre-trained to interact with. The actual implementation may have been very janky but the idea itself is sound: If you want to interact with arbitrary services then providing one single protocol (ideally not JSONRPC but hey) with a standar…

MCP is invisible to the LLM, or should be, under normal use

Re: Stateless MCP has recaptured my interest

#142

Post by acka »

floren wrote: post #41
kristjansson wrote: post #19

Earlier quoted context omitted.

Also: your messages causing your LLM (harness) to run CLIs on your computer? charming, thrilling, great fun. other people’s messages causing your LLM to run CLIs on your (cloud) computer? terrifying, awful, sickening, no fun at all

The LLM can't tell the difference between your messages and theirs, however many times you say "No mistakes"

Perhaps it is time to implement the lessons learned from Perl's taint checking[1], but this time for AI agent harnesses instead[2].

[1] https://en.wikipedia.org/wiki/Taint_checking

[2] https://arxiv.org/html/2607.03423v1

Re: Stateless MCP has recaptured my interest

#143

Post by alansaber »

rutierut wrote: post #78

In this thread I’m seeing a couple of the typical skill arguments. Skills don’t pollute the context window and, skills are more flexible and composable. In this specific post the SQL example gets around this by being super low-level, inherently composable, and low context consumption itself. I’m not sure this works for everyone. Looking at the Linear MCP for example, they basically just want to expose an API surface…

Skills definitely can pollute the context window. They can be verbose. You can have a very long list of advertised skills. You can try and hack around with skill execution inside of subagents etc but there is always a trade-off.

Re: Stateless MCP has recaptured my interest

#145

Post by jappgar »

ThePhysicist wrote: post #99
ai_critic wrote: post #50

It is amazing that folks getting paid several hundred thousand dollars a year...reinvented RPC-over-HTTP/JSON. You too, fellow web developers, are smart enough to work at Anthropic. I would love to see a proper engineering post-mortem for how this happened.

Coming up with solutions isn't so difficult, getting everyone behind one is the hard part. Obviously easier if you're one of the two biggest AI companies in the (Western) world.

The funny thing is that Anthropic and its boosters here framed MCP as a brilliant development.

It's a terrible protocol with zero value beyond bigtech sponsorship. We're all worse off for it.

Re: Stateless MCP has recaptured my interest

#146

Post by joombaga »

brabel wrote: post #129
sippeangelo wrote: post #115

Earlier quoted context omitted.

None of this is true. The point of "just CLI" is that LLMs are infinitely more trained on working CLI tools. There doesn't need to be real CLI tools behind the harness, as long as the interface is CLI-like.

Show me then how to convert my MCP, including auth, behind a CLI like interface. I don’t think that’s even possible.

Can you link it?

Re: Stateless MCP has recaptured my interest

#147

Post by lbourdages »

blitzar wrote: post #121
ai_critic wrote: post #50

It is amazing that folks getting paid several hundred thousand dollars a year...reinvented RPC-over-HTTP/JSON. You too, fellow web developers, are smart enough to work at Anthropic. I would love to see a proper engineering post-mortem for how this happened.

It was a vibe protocol and with the hindsight glasses on it is obvious. Someone should have said to the llm coming up with it -- "that sounds like an API with extra steps ... how about we keep these things (tool descriptions) and get rid of all this irrelevant - this will look good in my promotion meeting - overengineering (everything else)" You are absolutely right is what it would have replied with.

OpenAPI already exists, you can serve it on an endpoint for tool descriptions. Why did we need MCP again?

Re: Stateless MCP has recaptured my interest

#148

Post by blitzar »

lbourdages wrote: post #147
blitzar wrote: post #121

Earlier quoted context omitted.

It was a vibe protocol and with the hindsight glasses on it is obvious. Someone should have said to the llm coming up with it -- "that sounds like an API with extra steps ... how about we keep these things (tool descriptions) and get rid of all this irrelevant - this will look good in my promotion meeting - overengineering (everything else)" You are absolutely right is what it would have replied with.

OpenAPI already exists, you can serve it on an endpoint for tool descriptions. Why did we need MCP again?

that would be tokens, vibes and guesses all the way down

Re: Stateless MCP has recaptured my interest

#150

Post by jappgar »

shermantanktop wrote: post #35
pjmlp wrote: post #28

Earlier quoted context omitted.

Something that anyone doing distributed systems knows after a few scars, stateless servers are always better, and stateful only if there is no way around it. I learnt this with Sun RPC and the whole "The network is the computer". Somehow this keeps having to be relearnt.

I keep encountering mistakes by people who are convinced that agentic interactions are nothing new, that all the decades of hard-won lessons about service API design are 100% relevant and nothing more is needed, that agents can be relegated to the role of just-another-caller and left to the dismal domain of client developers. It’s not true. It’s not totally untrue, but don’t hide your head in the sand.

And agentic enthusiasts apparently believe agents can do anything except REST.
Post reply on HN