creative director @LabyrnthCo - AI/Web3 video content creator, educational videos | member of @aiscwork
Joined December 2021
- Tweets168K
- Following4.9K
- Followers21.3K
- Likes18K
this is what i wanted from security tools
it tries the bug for real first and only tells you when it's an actual problem
no more digging through warnings that turn out to be nothing
and it follows the issue across all your linked repos too
🤝 Paid partnership
Crypto sold us freedom and then put our balances on a billboard.
Every trade, every size, every exit. Public by default, forever.
@fhenix just went live on mainnet, so I tested Genesis with real money to find out if that era is finally over.
Try now: genesis.fhenix.io
🤝 Paid partnership
i’m making presentations with claude now
here’s the setup:
install claude code
open it and run:
/plugin marketplace add anthropics/skills
/plugin install document-skills@anthropic-agent-skills
put your brief, logo, images and data in one folder
open claude code inside that folder
then use this prompt:
use the pptx skill. turn these files into a 10-slide presentation. show me the outline first, keep one idea per slide, cite every number, match the colors from my logo and export an editable pptx
for changes, just send the slide number:
slide 4: replace the paragraph with a chart and keep the source underneath
github.com/anthropics/skills
Biscuit retweeted
My new designer never sends me an invoice.
It’s Arti, the personal design agent inside Lovart 2.0 by @lovart_ai
I’ve tried a lot of AI design tools. Most of them wait for a prompt, give you one image, and forget you the next day.
Arti remembers my brand. I set one Delegated Task, and every morning my new posts are already waiting for me, in my colors and my style.
I didn’t expect an AI tool to save me this much time.
Try Arti for your own brand here 👇
lovart.ai/?sourceId=905054
#BuiltWithLovart #BrandIdentity
🤝 Paid partnership
My new designer never sends me an invoice.
It’s Arti, the personal design agent inside Lovart 2.0 by @lovart_ai
I’ve tried a lot of AI design tools. Most of them wait for a prompt, give you one image, and forget you the next day.
Arti remembers my brand. I set one Delegated Task, and every morning my new posts are already waiting for me, in my colors and my style.
I didn’t expect an AI tool to save me this much time.
Try Arti for your own brand here 👇
lovart.ai/?sourceId=905054
#BuiltWithLovart #BrandIdentity
🤝 Paid partnership
this is how agents lose users
the answer can be good, but nobody wants to watch their quota disappear while the model keeps opening more subagents
fixing that will help more than another benchmark win
We're sorry for the rate limits and unnecessary quota consumption on the K3.1 Pro subscription plan. These were our issues, and we should have fixed them sooner.
Today, every subscriber on every plan gets a one-time usage reset, restoring 25% of their monthly quota (usage below 25% resets to zero), equal to one week of usage. Your usage bar will reflect this adjustment by end of day PST.
We've also adjusted the plan's parameters and are reducing unnecessary subagent calls and output length, so your quota goes further.
Thank you for your patience while we put this right.
this is the first grok bot workflow i’d build as an x creator
create five bots called researcher, scout, fact checker, writer and editor. add them to one group chat, then paste this:
“every morning at 9 am, check the official accounts and changelogs of openai, anthropic, xai, google deepmind and meta. only use updates published in the last 24 hours.
find the strongest story and its original source. verify every number and claim. find the best post to quote. write one draft in my tone, then remove repetition and anything that sounds promotional.
send me the source, verification notes and final draft. never publish without my approval.”
save it as a daily routine and the research will be waiting when you open x
Biscuit retweeted
i’m editing my own videos with claude now
here’s the setup:
1. install claude code
2. paste this into claude:
set up github.com/browser-use/video… for me. read install.md first and tell me when it’s ready
3. add your elevenlabs api key when asked
4. put your raw clips in one folder
5. open claude code inside that folder
then use this prompt:
remove bad takes and long pauses, keep the natural rhythm, add captions in my style and export 9:16
claude transcribes the clips and shows you a cut plan
once you approve it, the finished video appears as edit/final.mp4
for changes, send the timestamp:
undo the cut at 00:18
i’m editing my own videos with claude now
here’s the setup:
1. install claude code
2. paste this into claude:
set up github.com/browser-use/video… for me. read install.md first and tell me when it’s ready
3. add your elevenlabs api key when asked
4. put your raw clips in one folder
5. open claude code inside that folder
then use this prompt:
remove bad takes and long pauses, keep the natural rhythm, add captions in my style and export 9:16
claude transcribes the clips and shows you a cut plan
once you approve it, the finished video appears as edit/final.mp4
for changes, send the timestamp:
undo the cut at 00:18
why grok bot could be unstoppable from now on
the biggest advantage in ai may not be intelligence
it may be laziness
once one bot already knows you, setting up another from scratch feels like work
people rarely leave the tool that already knows them
the biggest ai agent mistake happens before it takes a single action
a vague goal, broad access and no stop condition
before i run one, i define:
→ the exact outcome
→ which tools and files it can touch
→ which actions need my approval
→ what proof it must return
then one final rule:
“if you cannot verify it, do not do it”
last year it took me 4 months and about 10 phone calls to see a dermatologist
- the first office wasn't taking new patients
- the second was out of network
- the third needed a referral
my primary doctor wanted to see me before sending one. their next opening was 6 weeks away
every answer came with another thing i had to arrange
that's the part of healthcare Vene is built to help with. coordinating referrals between doctors, booking appointments and handling the paperwork and insurance follow-ups
it can't make a waitlist shorter. but i would've happily handed over the work of getting everything ready so i could actually get on it.
🤝 Paid partnership
Today, I'm launching Vene: your AI advocate for healthcare.
I used to spend a dozen hours a week managing care for my dad, a disabled veteran, and my mom, who has a severe autoimmune disease. Healthcare was a full-time job for my family.
Our goal is to end the burden of care for every family:
When you're sick, or caring for someone who is, the hard part is everything after the hospital: the referral that never arrives, the test result nobody calls about, the bill that makes no sense, and everything in between.
For my dad, he spent hours on the phone to book an MRI. When he showed up, they told him to his face that insurance had stopped covering it. Then insurance told him to call a list of other offices and get the paperwork sent over himself, and he ended up giving up.
That’s why we’ve trained Vene to handle the toughest aspects of healthcare:
- Coordinating several specialists at once
- Challenging surprise bills and treatment denials
- Catching missed follow-ups and care gaps
My mom uses Vene to coordinate her primary care and endocrinologist, automatically refill and renew her prescriptions, and manage my little sister’s dental appointments.
They were my first users, and now over 5,000 families use Vene, and we're deploying it to another 50,000 patients across provider networks in the US.
the chain may soon be the least important detail
@fulldotfun puts robinhood chain, arc and bnb launches in one market
tokens start on a bonding curve and move to dex liquidity when it fills
when several chains share one interface, traders start choosing the platform before the network
Introducing full.fun 🚀
Create, launch, and trade memecoins across all chains.
The next era of on-chain trading starts here.
the model war just became invisible
most employees will not choose between grok, claude or chatgpt
they will use whichever model is already inside the software doing their work
ai coding has created a testing crisis
teams can now ship more code than they can understand
the worst bugs appear when an old job, a new deploy and a payment timeout collide in the wrong order
@brekenai runs those failures before users become the test suite
1/ Today we're introducing Breken, and with it the final job in software: production engineering.
Your tests check what your team thought of. Production checks everything else. Today, your users run that second test for you.
Breken runs it instead. Your real software, in worlds that behave like production, down to the chaos and the users, finding and fixing what breaks before a single user does.
Here it is: breken.ai