Posted by fg137 14 hours ago
Why not iterate on README.md with something like PARSEME.md?
AGENTS.md is not "structured format", then it'd be JSON or YAML or some other fucked up format instead. Currently at least, it's fully freeform and you can put whatever there, agents will do there best to follow it.
> determine which parts of the text are code and which aren't
With this argument you'd could claim ASCII/plaintext is "structured data" too, just group code with "===" and that's evident. Obviously that doesn't make a format "structured data".
Well, it was semantic from the beginning, the whole discussion I raised is because you seem unable to understand what "structured data" actually means in reality. I know perfectly fine what you meant, and you were misinformed about the meaning, it's OK to just say "Yeah, you're right" or just don't reply at all instead of double-down on not understand the meaning of words. Anyways, enjoy your day.
Excellent.
echo "Get stuffed, you aren't welcome." > AGENTS.md
cp AGENTS.md CLAUDE.md
Humans don't need it. Nor do they need to be reminded to launch subagents for simple tasks for example - which is another command that belongs in AGENTS.md.
For example, "don't write comments that reference things that you removed from the code" is commonly needed for coding agente and almost never needed for people.
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...
I've been deeply distrustful of Anthropic from early early days. They have always been openly disdainful of user feedback. I would not be surprised if later they try to implement more shenanigans to keep people locked into CC.
It's like the PMs for Claude Code are reward hacking their own reinforcement learning.
I have been happy with Claude Code lately, but I haven't explored other options much in over a year. Curious to try something else out if it's less rambling and tangential.
This is not necessarily a problem with the harness (IE, swapping from Claude Code wouldn't necessarily fix this).
If what you're looking to avoid is specifically 'rambling and tangential', you can get quite far with anything that adds directives to avoid those things, early in every context window. Ie, through use of Agents.md/claude.md, skills, hooks, and so on.
Changing model would also affect this. Fable 5, Opus 5, and Sonnet 5 are all going to average out to different levels of ramble. Openai, Xai, Google, etc; different providers models will also have different levels of ramble.
What Claude Code does take away from you is some level of control over what makes it into the context window. The system prompt which claude code append to the beginning of every session of course has measurable ramble-affect.
I quite like the Pi harness, most in part because important goal with the approach behind it is "give the user as much control over what makes it into the cotext window as possible."
Codex TUI is also good. Haven't touched it since moving to Pi however. Again- harness isnt the big "stop rambling" thing to change, tho.
I use a custom harness, but I constantly hear good things about Pi.
EDIT: Checked out Pi and Oh-My-Pi. 250k LoC and 1.5M LoC respectively. Sheesh. How's that for minimalism...
Mine's 50. (Lines.)
https://gist.github.com/a-n-d-a-i/bd50aaa4bdb15f9a4cc8176ee3...
Vendored in my own LLM micro library (100 LoC) so it's 150.[0]
There's no parallelism or anything, but I use it for surgical edits and it's much faster and cheaper than the official harnesses for some reason. (Absence of sysprompt bigger than my repo probably helps there...)
[0] This one uses OpenRouter so you can use it with any model, but jerry rigged Codex sub version available on demand :)
---
I also have a ultra turbo bloated version (500 lines... need to strip it down a bit!) which has autorun files (for grep-based context injection), notification (frog croak when agent done, etc.) Watch this space!
(Edit: I failed too mention, using a different harness with a CC/Anthropic subscription bumps you up to the pay-per-token rate.)
However! I have seen projects which use CC under the hood, in order to get the subsidized rates.
Writing comments which get me to go look through anthropic documentation, and find friction I wasnt aware of (CC does not have an app-server), refreshes my frustration towards anthropic.
Is there any provider that doesn’t do this? That’s the one I want to support.
> TLDR: they are all probably doing it because they are banking on you not using all your tokens.
This is... close. They offer ways to leverage openai subscription directly for 3rd party app use. https://learn.chatgpt.com/docs/app-server
> Is there any provider that doesn’t do this? That’s the one I want to support.
Most are easy to share with symlinks and @agents.
Also cause OpenAI was cool with you calling their sub in an automated way, but Anthropic very much was not (they were banning people at the time).
It has a sort of first mover advantage, but I wouldn't be surprised if basically any harness is superior to that crap.
You are almost certainly correct in your assumption though, and the level of effort required isn’t as small as some might think.
claude -p "@refactor_hits.txt support either CLAUDE.md or AGENTS.md in each location. if they both exist, prefer CLAUDE.md and do not concatenate"
Anthropic can do everything to make more money, including supporting the department of war, screwing its users if necessary and so on
I think we’re going to see some aggressive attempts at lock in quite soon. What choice do they have?
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