Sabitlenmiş Tweet
Perthro
174 posts


If sites built in @framer supported user accounts and visitor logins, what would you build first?
English

Prompt your way off your old CMS.
We shipped a skill that makes migrating to Sanity much easier.
It treats the migration as an ETL project, so you can fix rough edges and get your content into the exact shape you need.
Free to use, and it knows many sources, from visual builders to headless CMSs.
skills.sh/sanity-io/agen…

English

@jurrehoutkamp @wiamxyz @framer @mutinycorp gpt 5.5 in both, codex just... doesn't make it as good for some reason
also really like the blue outlines that whatever the native agent is working on has, while i can't really tell what my external agent is working on without checking
English

@perthro_studio @wiamxyz @framer @mutinycorp what model are you using in codex and what model are you using in framer?
English
Perthro retweetledi

At @mutinycorp, the marketing team now uses the Framer Agent to build major site changes themselves, saving $6k per month on agency support. Read their full story below ↓
English

@perthro_studio @densfs_ @framer Not part of the current roadmap, though we may consider alternative approaches later!
English

@jurrehoutkamp @wiamxyz @framer @mutinycorp true, but why does Codex suck compared to the Agent inside of Framer? it can't create things of nearly the same quality
English

@perthro_studio @wiamxyz @framer @mutinycorp Hot take: credits are only expensive if you don’t value your own time.
English

@perthro_studio @framer @mutinycorp Point of the tweet is that the agency is saving 6k,
Like yeah credits are expensive but if you wanna compare to the okd days,
how much time you’d spend or outsourcing,
Credits become the cheapest thing to come to our industry
English

Week 13 of the #FramerChallenge
Last week of the challenge.
Current: $1,099.63 / $7,500
This week: +$156.53
Breakdown:
• Component sales: $9
• Affiliate commission: $68.53
• Template sale: $79
What I worked on:
• Released a new Framer component
• Almost finished the additional resources for Poolify
I failed the goal by far, and that’s okay 🙏
English

Week 12 of the #framerchallenge
I still can’t quite believe it. This week, I won the Framer Agent Hackathon (+10k), made great progress with commissions, and closed two new client projects. There’s still a long way to go tho, why I may not make it until next week.
This week: (+$1.426)
- $560 Client project
- $465 Agency work
- $401 Dub commissions
Total so far: $5.377/ $9.000
English

I think the Framer Marketplace is about to change in a big way.
With models like Fable 5, anyone can generate a decent website in minutes. That probably means generic templates are going to become much less valuable.
But I don't think templates are going away.
If anything, this pushes creators to build things AI can't easily replicate. Better design systems. Better branding. More original visuals. More attention to detail.
The bar just got higher.
The templates that stand out won't be the ones with the most pages. They'll be the ones that have a clear design direction and actually feel premium.
AI can build a website.
Making something people genuinely want to use is a different challenge.
I'm interested to see how this changes the marketplace over the next year.
English
Perthro retweetledi

Fable 5 is now available in Framer.
It’s the most proactive model we’ve tested, often going beyond the brief with stronger styles, reusable elements, and polished first results. It’s especially good at original design work, shaders, and subtle animations.
In our evals, Fable 5 scored 83% overall, ahead of Opus 4.8 at 77%, and led every model on design at 81%. It uses 2× the credits of GPT 5.5.

English









