Re:Evaluating AGENTS.md: are they helpful for coding agents?
by nielstron on Tuesday February 17, 2026 @02:04AM
(#47044540)
Exactly my thoughts... the model should just auto ingest README and CONTRIBUTING when started.
Re:Evaluating AGENTS.md: are they helpful for coding agents?
by rmnclmnt on Tuesday February 17, 2026 @05:05AM
(#47045632)
And that makes total sense. Honestly working since a few days with Opus 4.6, it really feels like a competent coworker, but need some explicit conventions to follow … exactly when onboarding a new IC! So i think there is a bright light to be seen: this will force having proper and explicit contribution rules and conventions, both for humans and robots
Re:Evaluating AGENTS.md: are they helpful for coding agents?
by delaminator on Tuesday February 17, 2026 @07:57AM
(#47047011)
You could have claude --init create this hook and then it gets into the context at start and resume
Or create it in some other way
{
"hookSpecificOutput": {
"hookEventName": "SessionStart",
"additionalContext": "<contents of your file here>"
}
}
Re:Evaluating AGENTS.md: are they helpful for coding agents?
by delaminator on Tuesday February 17, 2026 @10:44AM
(#47048741)
Unlike other content - what you put in here survives compacting
Re:Evaluating AGENTS.md: are they helpful for coding agents?
by gordonhart on Tuesday February 17, 2026 @08:50AM
(#47047463)
Exactly, it's the same documentation any contributor would need, just actually up-to-date and pared down to the essentials because it's "tested" continuously. If I were starting out on a new codebase, AGENTS.md is the first place I'd look to get my bearings.
Maybe I’m wrong but sure feels like we might soon drop all of this extra cruft for more rationale practices