Live data from Hacker News

MCP is dead?

quandri.io

411–420 of 444 posts

Re: MCP is dead?

#411
post #354
post #325

Earlier quoted context omitted.

Yes I believe it is preloaded (from a recent test with latest claude-code actually). But that's an issue with the harness not something that's mandated by the MCP protocol.

I like MCP, but most servers and harnesses just dump all of the tools into context each message. Also the default tutorials for everything around MCP say to dump tools this way. It’s really hard to determine which tools an agent will need on the fly. The best idea I’ve seen was to do a “search tools” tool which an agent would use to find relevant tools, then have an “execute tool” tool which did what it says. Alterna…

Agreed 100%. It's sad when the harness floods the context with a whole MCP response and then see the model dumping everything to disk to process it with a script. That's something the harness should do whenever the result is large: dump the json into a file and just say to the model: "your data is in /tmp/foo.json, it's very large so be careful with it", but I don't remember having ever seen that; maybe some harnesses do it? Depending on what the model is after it might want to filter for this or that, no need to load everything in the context.

Re: MCP is dead?

#412

Earlier quoted context omitted.

And significantly increases the price.

I'm paying the same $100/month that I was back when the context window was 200K, so, no.

In other words, you're on a monthly plan, haven't read the not-so-fine print, and not going over your monthly limit, so not paying by the token yet. When you do, you will be in for a big surprise. Look it up. You'll thank me later.

Re: MCP is dead?

#413
post #390

Earlier quoted context omitted.

Different levels. Yes, MCP is a hack that could have been carefully built on prior art, and it would have been better for it. Yes, MCP is capable of expressing that prior art, and you can do semantic web concepts even if the wire protocol looks different.

> Different levels. How are they different? > Yes, MCP is capable of expressing that prior art, and you can do semantic web concepts even if the wire protocol looks different. What are you talking about?!

> How are they different?

Take a look at Semantic Web[1]

SW is about data interchange and standardizing semantic representations… it relies on formal, structured taxonomies of All The Nouns and then All The Verbs For Those Nouns.

MCP is structured at the protocol level, but its whole value is the bridge from structured to unstructured: natural language agent intents to specific APIs, domain-specific data to fuzzy natural language.

Semantic Web only works if someone has clearly defined the domain you want to operate in. MCP can work without that kind of definition (an API or prompt needs to exist, but it doesn’t need to be modeled in any way similar to RDF)

> What are you talking about?!

For instance, MCP-to-SPARQL bridge[2] that lets agents query SW data.

1. https://www.w3.org/wiki/SemanticWebArchitecture

2. https://pypi.org/project/mcp-server-sparql/

Re: MCP is dead?

#414

Earlier quoted context omitted.

MCP allows you to easily separate API requests from their access tokens, so that the LLM only has access to the requests part. Giving an LLM CLI access removes all boundaries, anything goes. EDIT: to add an example: I have a personal claw agent that I only use CLI, I don't care. But I'm also building an agent inside a company product, and there we use MCP all the way.

You can easily do this with simple Unix `chmod x-rw` on a wrapper that makes HTTP requests, adding the access token as it does.

The point is that you still need to expose a shell or a wrapper around exec to launch those commands. That is t appropriate when I want to lock the model down to working entirely from a sandboxed environment and expose some basic tools which are not “entire posix user space”.

MCP servers are not dead, they’re just overused in contexts they don’t really make sense in.

Re: MCP is dead?

#416
old and inaccurate knowledge..

- Skills do take up space in the context. The name and description of every skill goes into the system prompt. You can't add unlimited skills without context pressure.

- MCP context spam is less of a big deal now that there is deferred tool loading.

- If you work with the agents long enough you'll run into situations where using bash CLI tools suck and an MCP works better.

- MCP is not dead and has never been dead, it's the right solution in certain cases.

Re: MCP is dead?

#417
post #5

I'll kick myself for not remembering, but there was a fantastic article which suggested that MCP works at org level when unified, safe, access to internal utility APIs need to be given to non-technical staff who do use internal agent tools. Codify your workflow(s) via skills and share across instances, anything that needs context aware API access should be mcp...

But what is the advantage of MCP compared to having the agents access the API directly?

MPC comes with built in documentation over an API. Thanks to this agents can use it "intuitively".

Re: MCP is dead?

#418
post #22

I run the team at OpenAI that's responsible for the ChatGPT App Store, Codex plugins, and all things MCP. The thing that all these "MCP is dead" posts are missing is that whether or not MCP is used as a transport protocol is actually completely irrelevant. The reason MCP isn't dead is because practically ~every company on the planet is building an MCP server. I know this because we interact with all of them. Most of…

Wouldn't it be easier if MCP supported vector embeddings as input/output? It would shift its integration from a Broker Agent to a Deterministic Semantic Router. That by itself would prevent context bloat.

Re: MCP is dead?

#419
post #390

Earlier quoted context omitted.

> Different levels. How are they different? > Yes, MCP is capable of expressing that prior art, and you can do semantic web concepts even if the wire protocol looks different. What are you talking about?!

> How are they different? Take a look at Semantic Web[1] SW is about data interchange and standardizing semantic representations… it relies on formal, structured taxonomies of All The Nouns and then All The Verbs For Those Nouns. MCP is structured at the protocol level, but its whole value is the bridge from structured to unstructured: natural language agent intents to specific APIs, domain-specific data to fuzzy nat…

Ah. I see what you mean. I was mixed up in a side conversation and didn't catch on to what you meant originally.

Re: MCP is dead?

#420
post #411
post #354

Earlier quoted context omitted.

I like MCP, but most servers and harnesses just dump all of the tools into context each message. Also the default tutorials for everything around MCP say to dump tools this way. It’s really hard to determine which tools an agent will need on the fly. The best idea I’ve seen was to do a “search tools” tool which an agent would use to find relevant tools, then have an “execute tool” tool which did what it says. Alterna…

Agreed 100%. It's sad when the harness floods the context with a whole MCP response and then see the model dumping everything to disk to process it with a script. That's something the harness should do whenever the result is large: dump the json into a file and just say to the model: "your data is in /tmp/foo.json, it's very large so be careful with it", but I don't remember having ever seen that; maybe some harnesse…

That’s a good idea too, I might steal it.
Post reply on HN