medbrane

>here's no mechanism for progressive disclosure of parts of the tool's abilities

In fact there is: https://platform.claude.com/docs/en/agents-and-tools/tool-us...

If the special tool search tool is available, then a client would not load the descriptions of the tools in advance, but only for the ones found via the search tool. But it's not widely supported yet.

Indeed, the reconnect behavior is described in the protocol and the server will simply resume from the requested sequence id.

But hey, your fridge is now more secure because blockchain (I'm not kidding, it's in the press release)

Indeed, the article would have been correct one year ago.

Now, modern LLM APIs do require the tools to be described outside the prompt [1]. This negates the whole article, although one bit where he's right is that it does not matter if those tools are MCP tools or local, the call to LLM looks the same.

[1] https://platform.openai.com/docs/guides/function-calling?api...

That was in 2022, before LLMs, and they "lost" as in they had to pay back $482 USD.

>one of the oldest Western traditions is to demonize Russia and Russian people

Would you blame them? Who cares is something good happens now in Russia while they are brutally murdering their neighbors?

Nobody cares whether Hitler was great at drawing or not.

>Broadly speaking, Russian historians are generally of the opinion that the Holodomor did not constitute a genocide. Among Ukrainian historians the general opinion is that it did constitute a genocide.

https://en.wikipedia.org/wiki/Holodomor_genocide_question

No support for long, streaming interactions, like a terminal

An endpoint implementing this protocol would describe the agent and its capabilities, including examples. So I guess you could index that and create a discovery service.

But MCP doesn't claim to address agent to agent communication, right?

That's dependent on the particular LLM one uses.

But it can be a json with the tool name and the payload for the tool.

But not for gpt3.5 family.

Indeed, a reduced prompt size is unlikely to compensate for 8x inference cost increase.

Also fine tuning OpenAI's black box with data generated using gpt4 does not really require a lot more skills than prompt engineering.

But serverside tracking without consent would still be illegal, right? GDPR does not make a difference between cookies and other mechanisms.

What you ask for is now possible using Next.js and the new React Server Side components. Check the link in the sibling comment tree.

Note that you have to configure that on each device separately.

They do a lot better than a hypothetical Facebook store, which Epic suit would make possible.

>but don't assume you will be protected from coronavirus 100%.

Washing hands does not protects you 100% so why bother?

>because they're concerned about the efficacy of tests other countries are running. That's rich. I understand that the US developed test had a lot more issues that the standard not-invented-here one.

And a lot of countries do the test twice, and only act on double positives or double negatives.

>doing nothing is not doing harm How can this be OK? People will spread it if they think they are not sick.

So you are saying that the opposite (a long position) is advisable? But that would make the current investors behavior a rational one, right?

After 5 hours you can put then in the case for charging:

> 5 minutes in the case provides around 1 hour of listening.

Together with the case Apple quotes 24h.

They also keep it after you eventually opt out.

They have this question in the page:

> If I participate in the program, is there any personal information in the data collected?

..then they proceed to tell you what they don't collect. You can read between the lines that yes, they do collect PII.