Sending a bunch of fresh grad children in to barricade themselves in government offices, demand access keys, and uplink all the data via illegal Starlink wasn't enough. Now we need to give the drooling nazi child-porn-bot creator to our secure classified networks.
They really are just trying to tear this whole nation apart as brutally & as fast as possible, lend quarter to every bad power in the world & idolize the dirtbags all. Vice signalling until there's nothing left.
I mean, it looks like ordered Hegseth both war crimes and murders and America is not even at war. He is a guy who thinks bullying is being masculine and killing people due to feeling bad is how you prove that men more rational then women.
So, it makes 100% sense that he teams up with nazi billionaire to give him secrets in exchange of allowing him to more effectively mistreat people.
The guy went to a conference for Republican Women and drugged and raped the conference organiser who stepped in to stop him sexually harassing two other women attendees.
And this was probably considered a plus when hiring him into the current government.
This is absurdly great. 2-3x what you'd pay for input for GLM-4.7 normally, but it's still 1/10th Opus cost, at absurd speeds. And GLM-4.7 is quite good!
Use it to write a ton of tickets then maybe send a cheap Cerebras model like GPT-OSS upon it, if you want to make it cheap. Or use Opus or whatever to build the plan then have Cerebras GLM-4.7 implement.
You can definitely keep tweaking. It's also helpful just to ask it about what your possible concerns are and it will tell you and explain what it did.
I spent a good chunk of 2025 long time being super super careful & specific, using mostly very very cheap DeepSeek and just leading it by the leash at every moment and studying the output. It still felt like a huge win. But with more recent models, I have trust that they are doing ok, and I'm better at asking some questions once the code is written to hone my understanding. And mostly I just trust it now! I don't have to look carefully and tweak to exacting standards, because I've seen it do a ton of good work & am careful in what I ask.
There's other tactics that help. Rather than stare carefully at the code, making sure you and the AI are both running the program frequently, have a rig to test what's under development (ideally I'm an integration test type of way, which it can help set-up!). And then having what good programmers have long had, good observability tools at their back. Be that great logging or ideally sweet tracing. We have such better tools to see the high level behavior of systems now. AI with some prompts to go there can be extremely good about helping enhance that view.
It is going to feel different. But there's a lot you can do to get much better loops.
Seems like he is ramping up for a presidential run. It's just so deflating thinking that the people might continue having such awful tepid good for the rich & few others people running the show on and on and on.
I'm always curious how folks do status detection. Here you use tmux capture-pane and detect off that! Whew! Simple & direct!
I've been really enjoying how OpenCode is so extensible, how you can make great plugins that can for example read the session.idle event & then go do whatever they want. That does require dropping in some config asking for the plugin, which takes some effort & requires a restart (but your session will be right there & you can continue). It's technically elegant imo, and nice that there is the extensibility.
But hard to beat using screen as a framebuffer & just reading it out, for doing absolutely whatever it is you could possibly want to do! For example you can also detect permissions prompts, which I don't believe there is an event for!
Haha I'm all about the KISS principle. I also set up a snapshot testing framework so that people can submit screenshots of any messed up status reports and I can easily add them to the test suite to make sure we fix any issues that someone sees.
The suppression & occupation of humans & their places for this sick & disgusting wanton use of violence & detestableness is beyond belief. We are infested by the very worst people.
I really hope we can make some Ratatui backends that are intended for use cases like this. Just dump the component tree, dump specific windows, whatever. https://ratatui.rs/concepts/backends/
In that case we should have some sort of UI test backends I guess? This mcp was more for generic use cases which will allow any TUI framework in any language to work.
I really like this idea! I wish I knew other data expression engines for js.
I feel like adding filtering languages into our http endpoints is one of those forever bespoke tasks. This is probably not the right form for tackling that problem, since it is a fairly complex query language & processor and doesn't cleanly map to something we'd use in a URL query string. But it makes me miss odata a little bit. And it makes me wish there were more visible popular options for data expression languages.
I also ran into https://github.com/philschmid/mcp-cli this morning, which similarly makes a cli out of MCPs. I had been working a little on my own, but here we are!
I see this as a very viable path to "code mode", to having ai write code instead of directly invoking tools. Having the AI write down is work allows nice refinement over time, iteration, in a way that it just calling all the tools directly doesnt yield up.
https://blog.cloudflare.com/code-mode/https://news.ycombinator.com/item?id=45399204
They really are just trying to tear this whole nation apart as brutally & as fast as possible, lend quarter to every bad power in the world & idolize the dirtbags all. Vice signalling until there's nothing left.
reply