Plainpaper docs

Plainpaper documentation

Plainpaper is the board where your AI agent does your marketing, and where you steer and approve it. These docs take you from a blank workspace to campaigns your agent researches, drafts, and hands off, with you in control the whole way.

What Plainpaper is

You talk to your own AI agent (Claude for most people) and the agent does the marketing work: research, audiences, plans, emails, ad creative. Plainpaper is the place that work lives. Instead of scrolling past in a chat window, everything the agent produces lands as a card on a shared board: a research card, an audience card, a finished email you can preview, an ad staged in a realistic Instagram frame.

The loop looks like this:

  • You brief your agent in plain language, in the chat you already use.
  • The agent reads the board: the campaign so far, your brand guidelines, the process it should follow.
  • It writes cards back, and you watch the board fill in live.
  • You review, comment, and approve. Nothing ships without your click.
  • The agent pushes approved work out through each platform's own connection: your email tool, Meta, wherever it's going. Plainpaper itself never sends anything.

Why a board beats a chat window

Three things have to be true before an agent is useful on a real campaign, and the board supplies all three:

Memory
The campaign lives on the board, not in a chat transcript. Open a fresh session tomorrow and your agent reads the whole campaign back in one step: every artifact, its status, and how it all connects.
Direction
A board arrives carrying a campaign process, the exact structure of each deliverable, and your brand's voice. Your agent shows up already briefed.
Control
Statuses, comments, and an approval gate decide what the agent may do on its own and what waits for a human click. Anything that spends money or can't be undone always waits.

What Plainpaper deliberately doesn't do

Knowing the boundaries makes the rest easier to trust:

  • It's not an AI. You bring your own agent: Claude first, and any other agent that supports connections over MCP (the open standard that lets AI assistants talk to tools). Plainpaper never resells you a model.
  • It never sends. Plainpaper holds no email-platform or ad-account credentials, so it can never spend your money or send in your name. Your agent executes through each platform's own connection, after you approve.
  • Your assets stay private. Images and video on your boards live in private storage, so there are no public URLs for anyone to stumble on.

Where to start

How these docs are organized

  • Start here gets you connected and productive in one sitting.
  • The product explains each part of Plainpaper (boards, cards, deliverables, assets, approvals, guidelines, templates, and workspaces) and how your agent uses it.
  • How-to guides are task-shaped walkthroughs with prompts you can copy straight into your agent.
  • Resources hold the prompt library, plan details, and troubleshooting.
i

New to AI agents entirely? You don't need to be technical. If you can write a Slack message, you can brief an agent. The quickstart assumes nothing and gives you every prompt to paste.