BrandOS Builder · For agencies creating client work with AI
Turn client brand knowledge
into context AI can use.
Talk through a client’s brand with the AI Builder in your account. It asks focused questions and drafts a BrandOS alongside the conversation: one structured file covering the offer, audience and voice. You review each proposal before saving it.
One-time payment, before tax. Access by email after checkout. All sales are final.
From client knowledge to AI context
The brand knowledge is there.
Give it a structure AI can read.
The current offer is in a proposal. A language rule is buried in client feedback. Your account lead knows which claim was ruled out. Together, those decisions explain the brand. An AI session only has the parts someone gives it.
The Builder helps you bring that knowledge together and turn it into explicit context: the offer that is current, the audience it serves, the language to use and the claims the client can support.
What you already have
- Brand guidelines
- Current offer
- Client feedback
- Your team’s knowledge
BuilderExtract & structure.
You confirm the facts.
Format optimized for AI
- Audience
- Who this offer is for
- Offer
- What the client sells now
- Language
- Wording to use and avoid
- Claims
- What the sources support
- Open questions
- What still needs confirming
BrandOS is the context. The Builder helps you create it.
The Builder asks about one client’s brand, works with the context you supply and proposes a document you can review in your browser. The result is a reusable BRANDOS.md file. You confirm the facts, keep unresolved questions visible and load the relevant context into your AI tool.
Why I started writing
the context down.
At Brixon, we used AI with different context, instructions and setups. The same brief could produce different work depending on who ran it.
We tried ordinary folders and prompt libraries, then custom Claude agents. Those attempts exposed the same problem: the useful brand decisions still had to reach the person and session doing the work.
BrandOS grew out of that process. It gives the brand decisions a structure people can check and AI can read. The Builder takes you through preparing that context for a brand your agency works on.
I’m Christoph Sauerborn, AI Systems Engineer for Agencies. I built the Builder around that work: getting useful brand knowledge out of scattered material and into a form an AI session can use.
See the difference · An illustrative example
One missing condition.
A different promise to the client.
Take Nordfeld, a fictional ERP migration consultancy. It offers a data-model audit that takes two weeks from data access. That last condition changes what the page can promise.
# The decision in BrandOSThe audit takes two weeks from data access.Whenever the timeframe is used, that condition belongs beside it.
Before · The condition is missing
Book today. In two weeks, you will know what is holding up your migration.
This starts the clock at purchase. The team may not even have the data yet.
After · The rule is applied
Get a written assessment of your migration problem within two weeks of giving us access to your data. Start with the data-model audit.
The timeframe and its condition stay together. The customer knows what they need to provide.
That is the kind of decision a BrandOS makes explicit. A reviewer can check whether the draft applies it, and the next session can start with the same rule instead of waiting for the same correction.
The files give AI the context you have checked and loaded. They do not make every output on-brand. A person still reviews the draft, and your team keeps each client’s files current.
One-time payment, before tax. Access by email after checkout. All sales are final.
An AI conversation · €9 one-time
Your first usable BrandOS
starts with one client.
The BrandOS Builder guides you through preparing reusable brand context from the material and decisions your agency already has.
The build covers
- Positioning, audience and current offers — the facts behind the work, checked against the client’s sources.
- Language and claim boundaries — the wording you reject, the reason and an approved alternative where one exists.
- Working rules and open questions — what the AI session should follow and what still needs a decision.
BrandOS Builder
€9one-time, net
What you receive
- A conversational AI Builder with a side-by-side BrandOS draft, kept separate for each client.
- Start with a conversation or bring text notes and Markdown sources. The Builder asks about gaps and suggests changes; you decide what to keep.
- Save your working draft in your account and export BRANDOS.md to paste or attach in the AI tools you already use. You can also import an existing file.
Access arrives by email after checkout. No call or appointment.
One payment for the Builder, with a limited AI allowance. You supply the context, verify each client’s information and review proposed changes. Your saved document stays editable and exportable when the AI allowance is used.
Prices exclude tax. Any applicable tax and the total are shown at checkout before you pay.
One-time payment, before tax. Access by email after checkout. All sales are final.
The Builder covers preparing and exporting brand context. Auto-Load adds the hosted connection, brand asset uploads and a loading Skill for 12 months. You can use your exported file without it. Team access and infrastructure setup remain separate work.
Before you buy
Do I start with a client or my own agency?
Start with one client whose material and brand decisions you can verify. Create a separate BrandOS for each client and keep their files and AI sessions separate.
You can use the same method for your own agency’s brand. It does not have to come first.
We already have brand guidelines. Can we use them?
Yes. Use the current guidelines as a source, alongside the offer, approved brief and relevant feedback. The work is deciding what is current, making the rules explicit and marking anything still unconfirmed.
Do I need code or a particular AI tool?
No code or terminal is needed. Build and export your BrandOS in the web app, then paste or attach it in ChatGPT, Claude or another tool that accepts text context. Check that the tool received the files before using the output.
Use only client material you are allowed to use in that environment. The Builder’s conversation uses Anthropic’s Claude through OpenRouter. The AI tool where you use your exported BrandOS is separate.
Does it load automatically for the whole team?
No. The Builder covers context preparation and manual use. You can share the files with an authorised colleague, but you still need to pass on the current version and load it in the right session.
Automatic loading, access controls and team deployment are outside this purchase.
How long does it take?
That depends on the state of the brand’s material and the decisions you still need to confirm. Start with the context one real draft needs. There is no fixed completion-time promise.
What happens after checkout?
Your access link arrives by email. Open the Builder in your account, work through one brand’s material, export the context and try it on a draft you can check. Nothing needs to be scheduled.
Will AI follow every rule?
No. It can still miss an instruction or invent a detail. Check each draft against the sources, record the remaining errors and keep a person responsible for approval.
Is there a refund?
No. All sales are final. The Builder is a €9 one-time purchase, before tax; it does not promise a particular improvement in your AI output.
Give the next client draft
a checked starting point.
Choose one brand, prepare the context and put it to work on a draft you can review.
One-time payment, before tax. Access by email after checkout. All sales are final.