Re: AGENTS.md and CLAUDE.md Are Concessions to Proprietary Services, Not Defenses Against Them
Eli Zaretskii <[email protected]> Wed, 05 Aug 2026 14:56:41 +0300
| Newsgroups | gmane.emacs.devel |
|---|---|
| Message-ID | <[email protected]> |
> Date: Wed, 05 Aug 2026 12:18:29 +0300 > From: Jean Louis <[email protected]> > CC: Yuri Khan <[email protected]>, Sean Whitton <[email protected]>, > Lynn Winebarger <[email protected]>, Eli Zaretskii <[email protected]>, > Jim Porter <[email protected]> > > Why would GNU project follow some closed and corrupted OpenAI, Google or Anthropic forced standard? It's not a standard at all. You have been pointed up-thread to a URL from which it is clear that this is an emerging standard. Here's that URL again: https://agents.md/ And yet you still keep claiming your unfounded claims that this is not a standard, that no LLMs except a couple support it, etc. Please stop spreading misinformation. You are entitled to your opinions (which, btw, IMNSHO were abundantly described already, so no need to reiterate), but please be fair and make a point of telling that (a) this is your opinion, not facts, and (b) that URL such as the one above says the opposite. > It's not a static file. It can be deleted or modified. > > That means anybody can anyway generate LLM code for Emacs. Which is why we ask contributors whether they used LLM-generated code. Again, we've been though this already, no need to reiterate