musti retweeted
Dario: Fuck Chinese models
Sam Altman: Fuck Chinese models
Jensen: Yeah, Let’s Fuck Chinese model
MCP, CLI… and now WebMCP 🛠️
Any page on ATTIM now basically tells your agent “here’s what you can do” and hands it the tools, without clicking around..
One improvement at a time ☑
1
3
102
It's been a month since attim.link is part of the Claude Startups program, and I've been utilizing it heavily.
Awesome program for solo-funded devs, and I am truly grateful for it 🙌
2
4
96
When a new model launches such as Fable 5.2, first thing I do is picking the most complex/planning-heavy ticket from my Linear board.
After completing THAT ticket and feeling the relief, I've checked my T3-code usage data. It seems like I've used almost $50 worth of Fable 5.2 already! 😅
I am not using any Astra since I'm Opus-pilled, only usage you're seeing is from my DevOps-Hermes agent 🙃 (Hermes's skill usage is still OP imo)
2
61
new instagram/youtube/tiktok page formats just dropped
you don't need a 1,000-word prompt for a great explainer video
introducing /style-sampling: five proven styles for any topic
your agent suggests the fit, then researches, writes, voices and edits
they got us 1M views last week
try today in OSS or our Studio
1
69
my phone reminded me this screenshot I took from a few years ago, didn't even know the sword was from the berserk.
some things always comes together some way
2
77
musti retweeted
POV: You try Codex Ultra mode
(143 agents have spawned and only written tests so far)
openai team for the last few days
Thank you to the 7M active users who are now using Codex and ChatGPT Work.
We have added a banked reset to everyone's account to celebrate the milestone. You can apply the reset in the desktop app or on web and it will replenish the weekly usage for you.
Have fun out there.
1
143
I've got 3 usage reset available on my codex
is it weird that it almost feels like a passive income
stacking my resets on purpose to use them on a new model release btw
1
69
if you're trying to understand or use ugc marketing for your products, I'm sharing a report that makes a detailed comparison of different ugc platform
i've compiled this page from the conversations with my hermes agent that uses all types of skills for enhancing its research
lucky-signal-uhnl.attim.link…
1
62
I miss the old Codex, generous-mode Codex
Burn through the whole task, finish the code Codex
I hate the new Codex, the quota-low Codex
The “try again later,” out-of-flow Codex
I miss the sweet Codex, ship in a week Codex
I gotta say, at that time I’d like to meet Codex💔
3
37
I am super comfortable with this agent stack for a while now:
- coding: opencode -> gpt 5.5 & kimi k2.6
- daily/personal: chatgpt projects
- devops/cronjobs: hermes -> gpt 5.5
total cost: chatgpt plus($20) + opencode go ($10)























