Re: AGENTS.md and CLAUDE.md Are Concessions to Proprietary Services, Not Defenses Against Them

On August 4, 2026 1:02:16 PM GMT+03:00, Yuri Khan <yuri=2Ev=2Ekhan@gmail=2E=
com> wrote:
>On Tue, Aug 4, 2026 at 4:31=E2=80=AFPM Jean Louis <bugs@gnu=2Esupport> wr=
ote:
>
>> Nevertheless you can prove the point if you find any standard on using =
agents=2Emd and if such standard isn't there who is it them benefiting?
>
>https://agents=2Emd/
>
>It is an emerging convention and it benefits teams where contributors
>do not have to agree on a single harness=2E

I didn't know it until now=2E Yet as something invented by OpenAI and Goog=
le I don't want to adopt it personally=2E

Technically we know it cannot prevent LLM generated code=2E


>Something that interfaces with an LLM but is not a harness is just a
>chat UI=2E One could, in principle, manually paste the contents of a
>file into a chat UI, give an instruction, get text back, and copy that
>text back into the file=2E That would be code generation without a
>harness=2E People do not, in general, do that=2E

I am surprised really=2E

I know now you use those harnesses but you should know I don't use it=2E A=
nd you can see Emacs packages if they have some harnesses=2E=2E=2E=20

I just can't agree on that how users will use harnesses and how that file =
benefits as it's contradictory to what I see and know from my experience an=
d from Emacs packages and various LLM software suitable to inspect and chan=
ge files by using LLM=2E

I am using llama=2Ecpp and Cherry Studio and Emacs packages and functions=
=2E No harnesses there found=2E=20


Jean