Assaf Dagan

Building an AI-positive branding consultancy.

[Assaf Dagan @assafdagan]

Speaking to organics in Porto about the new way to automate brand work at scale.

[Pedro Oliveira @pcbo]

OpenClaw Porto in session!

[Dan Peguine @danpeguine]

When I work on content, I ask my agent to show me what it would look like on every platform

I give it a screenshot of LI or x or whatever and tell it to place the content in it.

Super simple and effective to get the vibe and share for feedback before posting.

[Dan Peguine @danpeguine]

i've been doing exactly the same over the past couple of weeks.

i find that the html approach is an excellent way to communicate ideas that i want to get across quickly.

oftentimes it’s a standalone html so i just send it in whatsapp to my clients and i don't even need to host it anywhere.

i find it an excellent way to communicate new ideas and experiment with lots of different variations.

[Assaf Dagan @assafdagan]

Well I know @figma has this wonderful Claude MCP but I felt it could be tailored to me, and my projects: I am building an exoskeleton for designers, and frankly, it fits any subject matter expert.

A little snippet of how this works: I had a new idea for a webpage for a project, and needed to define the product, wireframe it, write texts, design and put images in.

All of these details live across live conversations, design systems, figma files, teams chats and shared drives.

with my new platform, I can have an agent run through all those channels, get the idea from me and design it. This isn't the final design, but the first time i touch the project is after it has made 60-70% of the way.

This way, as a designer, I IMPROVE the work and make it unique and better rather than build a project where i feel i need to defend 'the way it was'.

It's my solution to the 'Ikea effect' so I don't fall in love with something I made just because I made it. I get to make this better, which is what my job is anyway.

More on the age of the subject matter expert soon.

[Dan Peguine @danpeguine]

now my clippy is an expert at whatever I want to learn + it has a chat window so i can easily follow along the conversation

1. It uses @karpathy's llm wiki so that it has specialized knowledge

2. it has a chat window so that i can follow along the conversation and copy-paste suggestions from clicky

I can also type in any questions or comments if I don't feel like pushing to talk

I tested this on the new Claude Managed Agents (scraped anthropics docs for it and put it into the wiki)

and it works great!

this is going to be an amazing way to learn new things!

[Assaf Dagan @assafdagan]

@danpeguine Can’t wait to check it!!

[Dan Peguine @danpeguine]

Clicky is an amazing tool so I forked it and added "proactive tutor mode":

It watches your screen and teaches you *without you asking anything*.

Every time you stop using your keyboard or your cursor, it gives you instructions/feedback.

Watch it teach me Figma in real time.

Two additional changes in my fork:
- auto copy the llm response to clipboard, when it helps you with writing (it helped me write this tweet)
- it only takes a screenshot of the window you are focused on, reducing llm confusion

[Dan Peguine @danpeguine]

https://t.co/a3j0BEmFuH

[Assaf Dagan @assafdagan]

@writenicecode Love this ! fun

[Danit Peleg ⌐◨-◨ @danitpeleg3d]

I built an automated workflow for 3d printed fashion! My dream workflow for years!

Here is what it does:
- You can upload a file of the pattern making of your garment that you want to make.
- You upload a picture for the texture.
- The software will know how to generate and translate the photo into this flat illusion textile that looks so cool when printing.
- You can see the printer bed so you'll be able to check if your part fits the printer bed or not
- You’re able to see the simulation of the textile before you generate the G-code.
- When you generate the G-code you already have the files ready to be printed.

I vibe coded this for seven hours straight. I built it with my OpenClaw agent that I'm using through my WhatsApp, using audio messages and screenshots.

Let me know if this is interesting to you to try it out!

[Assaf Dagan @assafdagan]

hey! such a cool effect. just shows this to my CC and asked if we can run this on one of my illustrations.
Thanks @emilkowalski

[Emil Kowalski @emilkowalski]

Made this interaction of Linear’s dithered logo for our latest release.

It’s using canvas with a lot of tiny dots. There’s an invisible circle around the cursor.

Any dot inside it gets pushed outward with a cubic falloff so it’s gentle on the edges, and strong at the center.

[Assaf Dagan @assafdagan]

@TimHaldorsson @EspressioAI Very interesting. We're in LIs with @danpeguine lets connect irl

[Assaf Dagan @assafdagan]

the new 'new phone who dis' message

[Assaf Dagan @assafdagan]

One of my dreams as a brand agency owner was always to have a branded portal for my clients (SOWs, decks, brand books).

It never happened because it was too expensive, time consuming, real overhead.

Now it’s all agentic and looks amazing.

I have this all as custom-made skills that are perfect for MY needs as a brand agency.

My agents can put this up for any client in minutes.

Me 5 years ago (heck, me 3 months ago) would NOT believe me that this would be ever possible.

I am feeling so empowered words can’t describe

[Assaf Dagan @assafdagan]

@anvisha Looks great, will definitely try it soon!

[Assaf Dagan @assafdagan]

@growthflo Can’t wait to try these thank you! Any skill that helps transcend slop is welcomed - thank you for your work

[Assaf Dagan @assafdagan]

@growthflo This is definitely going into my tool box

[Assaf Dagan @assafdagan]

https://t.co/qmFFvuE1Ta

[Assaf Dagan @assafdagan]

We're building an AI brand agency. Today we're open-sourcing the first tool from our workflow.

When I ran my NYC brand agency, competitive visual research took hours.

Pulling website images, logos, social assets for dozens of competitors, just to map where our client sits in the landscape.

Now we do the same in a few minutes with:

/visual-research [brand name]

The agent pulls brand images from websites, agency portfolios, and socials. Analyzes visual identity : color palette, brand evolution, design language.

Outputs everything in a way you cal look at without hurting your eyes. Its for designers who are technologists: If you read this far, you are already one of us.

Link below: Point your agent at it and install. Works with CC, OpenClaw and co.

More skills from our daily brand work coming soon. Feedback and contributions welcome.