Evaluating AGENTS.md: are they helpful for coding agents?

Re:Evaluating AGENTS.md: are they helpful for coding agents? by bootsmann on Tuesday February 17, 2026 @03:40AM (#47045102)

This reads a lot like bargaining stage. If agentic AI makes me a 10 times more productive developer, surely a 4% improvement is barely worth the token cost.

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by koiueo on Tuesday February 17, 2026 @03:54AM (#47045189)

> If agentic AI makes me a 10 times more productive

I'm not sure what you are suggesting exactly, but wanted to highlight this humongous "if".

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by zorked on Tuesday February 17, 2026 @03:58AM (#47045220)

There is now this whole subculture of dedicated .md writers that won't go away with some evidence.

The illusion of power and authority that LLMs give has deep psychological effects.

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by croes on Tuesday February 17, 2026 @04:32AM (#47045416)

10x is that quantity or quality?

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by Ragnarork on Tuesday February 17, 2026 @04:36AM (#47045444)

Also, "perceived" or "real"?

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by thesmart on Wednesday February 25, 2026 @05:53AM (#47149929)

Both

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by staticassertion on Tuesday February 17, 2026 @04:40AM (#47045467)

If something makes you 10x as effective and then you improve that thing by 4%...

[ Parent | on HN ]

Re:Evaluating AGENTS.md: are they helpful for coding agents? by zero_k on Tuesday February 17, 2026 @04:44AM (#47045494)

It's not only about the token cost! It's also my TIME cost! Much-much more expensive than tokens, it turns out ;)

[ Parent | on HN ]