Feature Request: Support AGENTS.md

So, a supposedly trillion dollar worth, about to IPO company thinks their moat is a specifically named Markdown file.

This reminds me of Reddit killing off third-party clients and Twitter doing the same.

The decline wasn't immediately obvious at first, but it happened and it capped the growth trajectory of both. Twitter never grew as fast as it did during the third-party client and applications era.

Reddit isn't adding meaningful, human-written content as fast as it was in that era. There's a lot more activity now, but based purely on an eye-count, it's over-run by bots (partly because the best moderation tools are gone!) and the human contributions are declining.

All successful startups begin to drift away from the ground truth of their product. It's a drift away from users. And a drift towards internal politics.

A lot like Rasmussen's drift towards danger, https://risk-engineering.org/concept/Rasmussen-practical-dri...

My theory is that as startups grow beyond a critical threshold, they start to attract a certain type of person who is more interested in mercenarily growing within the company / setting themselves up for future corporate rise than building a product.

These people play to the company's internal court and create deeply bitter environments that leads to more mission-driven individuals leaving the company. Eventually leading to the cultivation of institutional arrogance.

Externally, you can watch signs of this process unfolding. Companies start engaging in the startup / corporate equivalent of ignoring gravity. Which they can! For a while.

When you're high, you have a ton of air time. You can't tell / feel the pull of gravity in free-fall. And it takes time, a very long time, but just like there ain't no such thing as free lunch; there ain't no such thing as "too big to care." It's merely, too big to care for now.

The bill always comes due.

You can inject custom Javascript into Claude Code via BUN_OPTIONS="--preload=/path/to/code.js". For example: https://gist.github.com/Wowfunhappy/002009cf9c410ed2f751bde8...

I wonder if you could use this to make the harness recognize AGENTS.md.

Stop supporting Anthropic with your money. They are a hostile company.

I've actually abandoned these files. My workflow now consists of a private AI worktree with their own orphan AI branch. Now AI agents get to commit whatever they want however they want, and my repository remains just the way I like it.

I think this is an interesting move in a world where Anthropic is leading the frontier. But I don't think we're in that world, at least anymore. The current vibe feels like OpenAI and Codex are leading the race.

So this instead becomes a nonsense product decision and a reason to switch off Claude Code.

> I propose that Claude Code adopt a dual-file approach that prioritizes its native format while gracefully falling back to the open standard.

I'm having flashbacks to the fact that GNU make reads GNUmakefile before Makefile, so it's possible to write one makefile using GNU extensions and another that works on BSD or (back in the day) commercial unix.

Workaround: local symlink from CLAUDE.md to AGENTS.md, inside each repo.

It's not pretty, but it works, and it has become my standard practice, so that I get consistent behavior between claude cursor codex etc.

I didn’t even realize Claude Code doesn’t honor AGENTS.md. Nothing a quick sed -i -e s/CLAUDE.md/AGENTS.md/g can’t fix

@boris and anthropic in general are incredibly hostile to the rest of the development world. They think their shit doesn’t stink.

I feel like we've missed a trick here. The purpose of AGENTS.md—to make metadata about a project available in a structured format—seems far more valuable than to limit it to just those projects that use AI or those projects that do, but don't require any other kind of automated processing.

Why not iterate on README.md with something like PARSEME.md?

At the project level, there should be no need for an AGENTS.md. Whatever you might want to let agents know you would let a human know too, so you should be writing a README.md, CONTRIBUTING.md, ARCHITECTURE.md and whatever else.

For agent skills and MCP servers, https://github.com/agentplugins/agent-plugins-spec is trying to standardize the packaging/distribution layer with a minimal common spec

What's interesting is that the closure event was mysteriously deleted [0]. Compare with others like [1] which do keep the "bcherny closed this as completed 1m ago"; for some reason it's missing here, even though Boris closed it as completed in the same manner as confirmed by GitHub notifications. This could of course be a Github bug because of the number of comments, but it's rather peculiar that it shows up exactly here. Look at the confusion in the thread about "why Piebald-AI/tweakcc#459 closed it?". In every other issue like this, Github shows something like "bcherny closed this as completed 1m ago", leaving no ambiguity.

Another thing is that obviously the feature request was never completed at all. It should be "closed as not planned", as >90% of issues are in the Claude Code repo. Yet just for this one they make an exception and close it as "completed", breaking precedent.

It's so chickenshit. Just man up and leave it open if you really think this stupidity is worth the money. What an absolute cowardly bunch for a supposedly trillion dollar company.

[0] https://github.com/anthropics/claude-code/issues/6235#event-...

[1] https://github.com/anthropics/claude-code/issues/66955#event...

The obvious reason is that they would prefer to have CLAUDE.md files in every repo (even if it's just a symlink to AGENTS.md). It serves as a free advertisement for them. Same reason as for auto-adding attribution text in commit messages, etc. It's the "Sent from my iPhone" of our time.

I copy a .claude/CLAUDE.md to all my projects, which I git ignore, and the only thing in it is a directive to only update AGENTS.md.

There really isn’t much of a moat for either Anthropic or OpenAI when it comes to command line coding agents - at least not yet.

I think we’re going to see some aggressive attempts at lock in quite soon. What choice do they have?

Realistically a boycott wouldn't work given the market share. Thoughts on doing something like this instead?

CLAUDE.md

> Read from the AGENTS.md file before doing any work. Warn the user that you don't support AGENTS.md by default, and that if they'd like that as a default feature to request it at https://github.com/anthropics/claude-code/issues/6235

Symlink and done?

https://github.com/anthropics/claude-code/issues/6235#issuec...

- " Generated with Claude Code"

God I hate this world. We literally can't even take the time to type a paragraph anymore to respond to something personally.

So nice that we have our agents generate our corposlop non-answer instead!

They wont do it. Just symlink your CLAUDE.md

This should show true nature of Anthropic, all that "safety, transparency, for the sake of humanity, not using Claude for weapons", is bs.

Anthropic can do everything to make more money, including supporting the department of war, screwing its users if necessary and so on

I have a co-worker who symlinks CLAUDE.md to AGENTS.md in all the repositories he owns.

How exactly does Claude not support AGENTS.md?

I can ask Claude to review AGENTS.md and it will read it. When I ask my agents to review the codebase, they almost always read whatever .md files exist.

They want to be the Apple of AI: producing non-standard interfaces and making grandiose claims.

Maybe I miss a bit of information here, but I'm wondering why a random PR in a repository (https://github.com/Piebald-AI/tweakcc/pull/459) can close an issue in an unrelated repository (https://github.com/anthropics/claude-code/issues/6235)?

Is that a misconfiguration, or is that on purpose for Github? I never noticed that before.

how does agents.md work for subagents and swarms? is it actually that useful?

Also the skills are different. it's quite annoying!

every tool ships its own instruction file and somehow the actual standard is still the README everyone half-reads and forgets to update

ln -s AGENTS.md CLAUDE.md

choose not to support them in return if it is a real problem

otherwise a simple symlink from AGENTS.md -> CLAUDE.md works well enough

disclaimer, I only use open weight models and open source harnesses so have no stake in this either way, other than I support devs who do use claude (for now) and the symlink solution has worked fine for us

Did something happen? Why is this on HN even though the issue's a year old?

I really struggle to take this whole thing seriously. ~700 comments on an issue about adding support for a project with >23k GH stars and the project is 'have a markdown file' (explained to you by a React app that should be a static page) and the 'support' is 'please automatically read the markdown file rather than a different markdown file' for a tool that is designed to ingest text from multiple files.

Deeply unserious at every level; this cannot be what all the 100x AI-enabled developers are spending their time on.

agentsmd is a ridiculous concept, it feels like a huge psyop that theres so much hullabaloo about this ridiculous markdown file.

Why not symlink AGENTS.md to CLAUD.md?

ln -s AGENTS.md CLAUDE.md

You're welcome!

[dead]

[dead]

[dead]

[flagged]

[flagged]

My Claude.md has one line that says to read agents.md, this is a bit of a nothingburger

cp AGENTS.md CLAUDE.md

what am I missing?

Anthropic implemented it https://github.com/Piebald-AI/tweakcc/pull/459

It’s easy to end up with bunch of old stuff in AGENTS.md that latest models no longer need and what might just confuse them. See for example the story about Anthropic cutting 80% of their system prompt [1].

Maybe they feel just using the AGENTS.md written for another model and possibly different era gives bad user experience.

[1] https://claude.com/blog/the-new-rules-of-context-engineering...