News for hackers. Stuff that matters.

A comment on Show HN: Agents.json – OpenAPI Specification for LLMs. Parent comment

Re:Show HN: Agents.json – OpenAPI Specification for LLMs by esafak on Monday March 03, 2025 @03:19PM (#43246193)
Isn't the idea to create a data lake to better inform models? Why are you bearish on stateful protocols? Could you elaborate on your thinking?
[ Parent | on HN ]
Re:Show HN: Agents.json – OpenAPI Specification for LLMs by yompal on Monday March 03, 2025 @04:49PM (#43247254)
Bearish on everyone needing to be on stateful protocols. Developers should have the option to have their state managed internal to their application.
[ Parent | on HN ]
Re:Show HN: Agents.json – OpenAPI Specification for LLMs by esafak on Monday March 03, 2025 @04:57PM (#43247332)
Can't you simply use a stateful protocol and not report any state? Doesn't statefulness subsume statelessness? I am beginning to wrap my head around this space, so excuse the naive questions.
[ Parent | on HN ]
Re:Show HN: Agents.json – OpenAPI Specification for LLMs by yompal on Monday March 03, 2025 @05:06PM (#43247429)
No worries! In other cases, I believe you would be right. But splitting up context is not optional with MCP. Part of the whole state will always reside in an external entity.
[ Parent | on HN ]