Found something good?

The New Workflow

Yesterday, Figma introduced Claude Code integration, where code flows to Figma and back again in a continuous loop. This is important because the loop works both ways; your team creates in Claude Code using your structured brand files, sends it to Figma, your designer refines it on the canvas, then brings it back to code with Figma MCP.

Certain files (like brand-positioning.yaml and messaging-framework.md) serve as the single source of truth for both sides - the agent reads them when generating code, and the designer references them when refining in Figma. They both employ the same positioning, same voice, and same structure.

However, this only works if your brand has structured data, not just Figma files.

At Little Plains, we’ve been preparing for this over the past few months. We've been slowly shifting from just deliver brand guidelines to delivering brand infrastructure. I think it's cool and hopefully this is helpful.

Here’s how we're approaching this:

Brand Infrastructure

We've been working on producing two layers for each brand:

It's less about replacing one with the other, and more about giving teams what they need, however they're working. Your designer needs Figma files and your agent needs structured data, and both need the same positioning, voice, and structure.

Let's review each:

First: The Human-Facing System

This is the craft: visual identity, typography, motion, website, narrative, and guidelines. PDFs, Figma files, design systems; this is what people see and feel. This is what we're 'used to' delivering, and receiving.

Second: The Agent-Readable System

This is the structure. Markdown, YAML, and JSON files that define positioning, voice, differentiation. It’s for your team to actually use when they're working with agents (which increasingly, we all are).

We call it an Agent-First Brand Kit, a structured knowledge system where every piece of brand guidance lives in a chunk (one of my favorite words ever), approximately 400 tokens each, with metadata that agents can parse and humans can read.

Here's a selection of what we’re giving select clients, using the same example (Baba Care) as above:

Brand Positioning

Persona Care

Retrieval Rules

Brand Guideline Note:

The full brand system typically includes 15-20 chunks (one of my favorite words in the space) across positioning, messaging, audience, identity, and guardrails. The three above are our foundation. What’s also included is below:

Positioning chunks:

Messaging chunks:

Audience chunks:

Identity chunks:

Guardrails chunks:

What We're Building: Practice What You Preach

This month, we’ve been rolling out providing our partners both a traditional brand system for humans and structured data for agents. Same positioning, voice, and values, just two formats. We call them dual-native brand systems, and think it will soon be a core part of all digital kits.

This is a byproduct of us needing to build systems that keep up with the speed of the teams we work with. Because the startups we collaborate with are on the front lines of building in the AI space, they need to ship fast, and that almost always means they’re using agents. They don't need the same brand kit that worked in 2020, they need kits that work today.

At Little Plains, we're running, and building, twelve agents. They are for each key area of our business, and will work alongside our team (the humans, and the other agents). Each one has identity, soul, and principles files. Each one has a distinct personality to help make them less generic assistants, and more teammates.

For example, here’s Donna, one of our latest teammates. We write this in markdown, give it to the agent, and the agent becomes Donna. She's not generic; she has a clear point of view. When we’re talking in Slack, you know it's her. That's what soul.md does: it turns a tool into a teammate.

Identity

Soul

Principles

When we shared this approach with @DanBatten, Head of AI at Infinite Garden (where he is responsible for creating an entire company run by agents), his response validated how fundamental this shift is:

“This is super smart. ...When done right, it actually feeds through the whole architecture. It becomes the foundation for how you think about the general knowledge and context base. Everything becomes these 400 token packets of info with the front matter so they can be very quickly indexed. Here is your human deliverable and here is your agent ready deliverable!"

This isn't just a new deliverable format - it's a new way of thinking about brand architecture itself. And that's fun!

The Peter Steinberger Principle

Last week @steipete, the founder of @openclaw, appeared on @lexfridman's podcast. During the (great) interview, I loved this line:

"AI doesn't replace craft. It makes craft more valuable. Because when everything can be generated instantly, the thing made with love is the one people remember."

That's it.

If your team is using AI to move faster, cool. But, if every output is generic because there's no structure to guide it, you're just speeding toward mediocrity. Your visual system? That's craft. Your motion language? That's craft. Your guidelines? That's craft. Your agent’s markdown files are how your team and their AI tools understand that craft. Both matter.

Leapfrogging

Little Plains started in early 2024, five years after my last agency, @ginlane was acqui-hired into the business it incubated, Pattern Brands. In that half decade, a lot had changed.

Starting anew was scary. I remember wondering how the heck I was going to catch up to a space I'd been out of for so long.

Sitting around one day, for whatever reason, I recalled a concept I read about, ‘leapfrogging technology.’

The specific example I read about was in the late ‘90s, when many emerging countries had little pre-existing telecom infrastructure. Most never built dense copper phone networks because they were too costly and slow. When mobile got cheap and robust, these countries didn't replace a legacy system; they leapfrogged straight to cell towers.

That was the lightbulb moment I needed...

I thought of skipping the traditional agency methods of the last era to focus fully on going AI-first. I didn't know what that meant, but I don't think I had a choice. I still feel the same today. However, as Littlefinger said in Game of Thrones, 'Chaos is a ladder’ and that moment felt chaotically opportunistic.

I bring all this up because how fast everything is moving today can also feel scary. I think if you hold on to the past too much, it is a bit much! But, if you allow yourself to be curious, to learn, to experiment, and to explore, you might find the opportunity in the chaos to leapfrog forward yourself. Here's hoping so ~

If you want to create and innovate, hit us up!