Privacy and analytics
We use necessary technical cookies and, only with your consent, analytics and marketing pixels to improve the website and measure campaigns.
Connect your brand to Claude or ChatGPT and read its Genome, Kit, Library and work from there. Each account authorises which brands and which permissions, and nothing is published or changed without a person approving it in Nömad.
The essentials
Brand OS context available outside Brand OS, with your permission and under your control.
How it works
A flow designed so control always stays with your account.
Capabilities
Reading your brand context, and proposals a person always approves.
In detail
FAQ
Model Context Protocol, an open standard for AI assistants to connect to external data and tools. Nömad uses it so your assistant works with your brand's real context.
Brand OS
Discover the rest of Brand OS features and how they fit together.
Yes. The assistant connects with your account and respects your role in each brand; without an account there is no access to the context.
Only what you authorise: the context of the brands you choose and the status of their work, never more than your role allows. Every query is logged and you can remove access whenever you want.
No. It can prepare post drafts and ideas and request a review, but publishing, charging or changing a connection is always decided by a person in Nömad.
Yes. The server runs at https://app.nomad.ooo/api/mcp.
More and more work happens inside an AI assistant. With the Nömad connector, that assistant works with your brand's real context and without copy and paste: you connect it once with your account and, from then on, Claude, ChatGPT or another compatible assistant can read the Genome, Brand Kit, Library and work of the brands you choose.
MCP (Model Context Protocol) is the open standard AI assistants use to connect to external data and tools. Nömad runs its MCP server at https://app.nomad.ooo/api/mcp: you add it to your assistant, sign in with your Nömad account and choose which brands and which data it can reach.
The connection is personal. Each person authorises their assistant with their own account, and access never exceeds what their role allows in each brand. You can remove access to one brand at any time without touching the others, and every query is logged.
Almost everything the assistant does is read: the Genome, the Kit, the Library and the status of work. The only things it can write are proposals: post drafts, ideas and review requests. It never publishes, charges, changes a connection or grants anyone access. A person makes those decisions in Nömad.
Assistants that support it can read your brand as files: a brand summary, a design guide and its design tokens (colours and typefaces) in the W3C standard format.
If you prefer to build it yourself, you can create API keys with specific permissions and revoke them whenever you want. The capabilities the assistant uses are available over HTTP at https://app.nomad.ooo/api/v1.
