About Plainpaper
Plainpaper is a small company in Amsterdam building one thing: the shared canvas where a marketer and an AI agent do campaign work together, in a place where both of them can see it.
What Plainpaper is
Plainpaper is a board. Your AI agent writes onto it, and you steer and approve what lands there. Connect Claude, ChatGPT, Cursor or any other client that speaks MCP, and the agent reads the board's current state, does the work, and writes it back as typed cards: research, segments, insights, plans, emails, ad creative, results. A campaign stops being a scroll of chat messages and becomes something you can look at.
The board carries your rules with it. A guidelines library holds your brand, voice and do-and-don't list, and the agent has to read it before it authors anything, so the tenth card sounds like the first. Nothing leaves the board until you approve it.
Why we built it
AI agents got good at marketing work faster than anyone built a decent place to put that work. The output ends up buried in chat logs: impossible to review as a whole, impossible to hand to a colleague, impossible to pick up again three weeks later. Every campaign starts by re-explaining the brand to a model that has already forgotten it.
Docs and whiteboards do not fix this either, because an agent cannot really work in them. We wanted the surface itself to be the interface both sides use: you see a board, the agent sees structured cards it can read and revise, and it is the same object.
What Plainpaper deliberately is not
Plainpaper is not an LLM. We do not train, host or resell models, so there is no model lock-in and no token markup. You bring the assistant you already pay for, and you can swap it tomorrow.
Plainpaper is also not the execution layer. We hold no credentials for the platforms you market on, and there is no tool here that sends an email, publishes an ad, or deletes anything on an outside service. When approved work ships, it ships through that platform's own integration, started by your agent after you said yes. The security page spells out where that boundary sits and why.
Where the name comes from
A plain sheet of paper is the least opinionated tool in any office. It does not ask what template you want or which of eleven views you meant. You put something on it and everyone at the table can see the same thing.
That is the standard we hold the product to, and it is why we write Plainpaper as one word: it is a name, not a description of stationery.
Who we are and where things live
We are a small team in Amsterdam, shipping weekly. There is no support tier and no chatbot in front of us: [email protected] reaches the people who write the code, and we aim to answer within one business day.
Your account, your content and your uploaded files are stored in the EU. Card details are handled by Stripe and never touch our systems. The full list of processors, what each one does and where it runs, is in the privacy policy. The agent skills we publish are open on GitHub.
Trying it
The fastest way to understand Plainpaper is to watch an agent fill a board. The quickstart takes about ten minutes end to end, and the playbook library gives you a real campaign structure to start from instead of an empty canvas.
Keep reading
- Plainpaper vs a blank chat: why a board beats a transcript, a whiteboard and a doc.
- Plainpaper docs: the product documentation, from first board to full campaign.
- Security: where data lives, how workspaces are isolated, how to report a problem.
- Support: how to reach us and what to include so the first reply solves it.