asar

thanks! it now clicked for me. so instead of cat its read_file, even if read_file resolves to cat, cat is not always available.

cool solution to a problem that shouldnt even exist I'd say. how did it become ok to be recorded at all times without prior consent??

This to me looks like it belongs into the harness, not the interface.

Working on a mobile interface for terminal multiplexers like tmux/zellij/herdr. This focuses a lot on my need to manage agents on the go, it's been quite helpful to myself so far.

MIT licensed and available on GitHub

https://github.com/AltanS/collie

On Twitter they mentioned that it was unfortunate timing as the 3.7 flash release collided with ox alpha.

I only saw this now, thanks for writing this up. Will take a look and address these in one of the upcoming releases.

I think the reason is that after all the benchmark comparisons, anthropic models still perform much more reliably and better in day to day use. Every day now there is a new supposedly fable level model that then somehow completely vanishes in a couple of days. While fable is still the best model and if only it were cheaper less people would be hating I suppose.

This really reminds of the spec comparison obsession android users had in the 2010s. Every 6months an 'iPhone killer' was released, but they were only ever superior in total isolation.

if you're using herdr, i built a plugin exactly for this reason. its a pwa so requires a proper frontdoor for secured access (tailscale, netbird etc.).

https://github.com/AltanS/collie

I've been working on collie[1] a plugin for herdr that runs a web app on the herdr host which can then be served via tailscale for example. This has made accessing my agents on the go a lot easier, paired with session persistence from herdr it's been quite nice to work with.

[1] https://github.com/AltanS/collie

Love what you're doing at cactus, keep it up!!

They should opt out of Google. What they are doing here in open daylight is insane. They stole the entire internet to train their models and are now shoving AI overviews down everyones throat. Clicks to publishers were already decreasing over the last years even without AI, but 2026 is showing what the future will look like, zero clicks from Google, because they dont care. Google got what it wanted and is now racing to AGI spending all its cash on datacenters, not caring about all the casualties along the way. Google, on its current trajectory, is truly evil in my opinion.

I'm really not a fan of regulation, but someone has to step in and have Google seperate its products, search should be free of AI overviews imo.

In / Out Price

$0,10 / $0,20per 1M

from openrouter

really cool! didn't know this existed, thanks! Going to have to figure out how to make this work with headscale though, looks a bit tedious.

So I guess we (and a lot of other people) have had the same problem, which is managing your agents on the go. I decided to build a plugin[1] for my terminal multiplexer (herdr) to access the sessions via PWA served through a tailnet.

[1] https://github.com/AltanS/collie

this looks nice! i've been using herdr the last couple of weeks as a terminal multiplexer for agents, which works amazingly well.

$1.5/m input tokens $9/m output tokens

6x the price of 3.1 flash lite

The model is (like Composer 2) based on Kimi K2.5 and they claim SOTA performance for 1/10th of the cost. The tweet also mentions that they've started a new model from scratch on Colossus 2 (xAI/SpaceX Cluster). Really impressive how they've made this jump from being called the vscode fork with no moat just a couple of months ago.

Always thought of this as two cars driving faster than you on the road. After a certain distance it's clear both are faster than you, but really hard to say which one is the fastest.

In the same boat and ready to downgrade. But this must be on their radar, or they were/are losing money with opus...

I see, thanks for explaining and congrats on the launch! After re-reading the description, the ability to use other frameworks might become a USP too.

Just a random remark, what's annoying and a pain point in my workflow are definitely proper development environments for agents . Not just runtimes but also managing secrets etc. Maybe an avenue to explore and use in marketing copy.

I agree with the general sentiment here that this is the future of coding for a lot of tasks. But in terms of a business case for your product I'm really struggling to see how this beats Claude code action? Which integrates directly with GitHub, at no additional cost, and I can use an oauth token to use my subscription.

Haven't used Windsurf yet, but in other tools this is called 'Agent' mode. So you open up the chat modal to talk to an LLM, then select 'Agent' mode and send your prompt.

Yes, they've updated the docs since last week, I guess. Before, it didn't mention the 20% markup.

thanks! i never set this up properly. did it now though, really cool!

Cursor recently lost me as a customer. Too many updates that disturb my workflow and productivity, no easy way to roll back versions, super sparse changelogs, lots of magic in context building, really untransparent pricing on max mode. I recently made the switch to Claude Code on the Max plan and I couldn't be happier. The only real thing I'm missing is the diff view across files, but I assume it's just a matter of time until that's properly implemented in Zed or VSCode.

Amazing, really excited to try this out. And thanks for the time you took to write this up!

Personally, I've been using Cursor since day 1. Lately with Gemini 2.5 Pro. I've also started experimenting with Zed and local models served via ollama in the last couple of days. Unfortunately, without good results so far.

I've created a list of self-hostable alternatives to cursor that I try to keep updated. https://selfhostedworld.com/alternative/cursor/

This sounds really cool. Can you explain your workflow in a bit more detail? i.e. how exactly you work with codex to implement features, fix bugs etc.

I love this analogy.

Really interesting method. I've been calling a similar strategy the library effect. Whenever I work in an environment where other people are productive (or at least look productive) I can focus much better and get in the zone. It's now gotten to a point where I'm actively seeking desks with my screen exposed to the room, so people would be able to see me procrastinate, guilt tripping me to limit this sort of behavior.