K.

Building — vision protocol for AI agents. Give your agent eyes. SRE → founder @acctixodigital

[K. @Boredktht]

Wait, Apple container is out and nobody is talking about it? Are we finally about to get rid of the Docker monolith on Macs?

[K. @Boredktht]

Honestly, I’m still shocked this works. Design QA is starting to feel like something we can actually automate.

[K. @Boredktht]

mdvp-cli is now CI-ready: run design audits in CI/CD, GitHub Actions, pre-commit hooks, or local dev loops, with audit runtime improved by over 3x

[K. @Boredktht]

@wesbos @grok, would https://t.co/BeJoh5aD79 (https://t.co/jFuHfMKexk) might be useful to detect and then fix that?

[K. @Boredktht]

Also, Codex environments save a ton of usage limits because the model is much less likely to get stuck in command iteration loops, for example -deploys??.

Computer Use works absolutely excellently.

And I still have not seen anything like this from the Anthropic side.

[K. @Boredktht]

codex pro for 100$ not rly bad tbh.

10% of weekly usage over 7 hours
5.5 on medium and high is way better than opus 4.6 max...

happy i can actually touch grass unlike claude bc i can continue any chat with codex remote (no need to have it enabled for each chat)

thanks for delivering 🙏 @sama @OpenAI

[K. @Boredktht]

This works better than Claude design

[K. @Boredktht]

i'm completely open-sourcing mdvp/cli tool. use it free forever, contributions welcome.

[FlyMy.AI @FlyMy_AI]

🚀Today we ship @FlyMy_AI Agents.
The world's first all-in-one agentic cloud.
The modern way to build, integrate, and scale production AI agents.

3 steps to a production agent:

1. Connect your work tools to FlyMy
2. Describe what the agent should do - in text or 5 lines of code
3. Set execution rules: manual, scheduled, or integrated into your backend
4. Done! Agent works on scale!

Everything in one place: 800+ MCPs, hundreds of AI models, brain, memory, sandboxes.

Stop building from scratch. Stop waiting for infra. Compress 6 months into a day.

#1 on @ArtificialAnlys benchmarks. Stable, secure, scalable from day one.
Try https://t.co/FxgVfIzDiM →

[K. @Boredktht]

@mitchellh Can someone please tell me why we need Ghostty-level customizable terminals for?
Or Hyprland.
You’re literally configuring your tools instead of using them

[K. @Boredktht]

@dunkhippo33 multi agent demos always look magical right until prioritization shows up

[K. @Boredktht]

@johncoogan @sama openai buying distribution was inevitable
models are getting commoditized. audience isn't

[K. @Boredktht]

@cursor_ai everyone is focused on more generation
feels like the real bottleneck is review now

[K. @Boredktht]

@dhruvtwt_ it'd be wild if Anthropic shipped fixes they found here first

[K. @Boredktht]

@browser_use finally someone solved the auth problem for headless agents

[K. @Boredktht]

Maybe we need to hire someone in Palo Alto to tape flyers about agent sandboxes or x402

[K. @Boredktht]

@Google native function-calling for agentic workflows is the part that actually matters here
everything else is benchmarks, but this is the thing that changes what you can build locally