Agents write fast. You decide what ships.

PMKIN is where your team and your AI agents plan, write and review content together. Every draft starts from a brief and gets a review before it reaches your site.

Works with Claude, ChatGPT, Cursor and any MCP client.

The problem

AI made writing fast. Everything around it got messier.

Drafts now come from people and agents, in docs, chats and CMS fields. Without one place to run it all, the same questions come up every week.

No one knows what’s in progress.
Ideas live in docs, chats and someone’s head, so two people, or two agents, end up writing the same post.
AI drafts start from nothing.
An agent doesn’t know your keyword, your audience or what you’ve already published, so the draft comes out generic.
Nothing checks a draft before it ships.
Text gets pasted from a chat window into the CMS, with no review and no record of who wrote what.
You guess what to write next.
Search data sits in another tool, so the next post is a guess.

How PMKIN solves it

One workflow for your team and your agents.

Every piece of content takes the same path, whether a person or an agent writes it.

  1. Plan in one place

    Ideas, drafts and due dates sit on one shared planner, so everyone can see where each post is.

    Answers “Did the webhooks guide ever get published?”

  2. Start from a brief

    A brief holds the keyword, audience and outline. One teammate or agent claims it and writes from it.

    Answers “Who’s writing the pricing post?”

  3. Review before it ships

    Every draft lands in a review queue. Approve it, or send it back with notes the agent reads.

    Answers “Can someone check it before it goes live?”

  4. Learn what works

    Search Console data shows what readers look for and where you’re close to page one.

    Answers “What are we writing next month?”

Content planning

See what's coming, what's stuck and what's still missing.

Every document lives on a board with a column per status, and on a calendar by its due date. Drag a card to move it along.

Group related posts into topic clusters, and PMKIN shows how far each topic is, like “3 of 8 published”.

Briefs

Write the brief once. Hand it to anyone.

A brief says what to write: the target keyword, the audience, the outline and the posts to link to. Fill in what helps.

Mark it ready and a teammate or an agent claims it. Only one can, so two agents never write the same post.

MCP server

Your agents join the team, not the other way around.

Connect Claude, ChatGPT, Cursor or any MCP client in a couple of minutes. Agents read your project, plan ideas, claim briefs, write drafts, add images and send their work to review.

They can't touch API keys, members or billing. And every change they make shows which agent made it, and for whom.

  • Claude
  • ChatGPT
  • Cursor
  • Claude Code

Reviews

Nothing goes live without a second pair of eyes.

Drafts from people and agents land in the same review queue, with a note for the reviewer. Approve to publish, or request changes and say what to fix.

  • Agents read your feedback before they revise
  • Every version is saved, so you can always go back
  • An activity log of everything agents did

SEO data

Know what to write next.

Connect Google Search Console and PMKIN imports your clicks, impressions and rankings every day, matched to your documents.

It points out searches where you rank just off page one, and suggests real queries as keywords when you write a brief. Your agents can read the same data.

For developers

A headless CMS your website reads in one request.

Your site fetches published documents over the GraphQL delivery API and renders them however you like: a blog, docs or a changelog. Drafts, briefs and reviews stay inside PMKIN.

Automate the rest with the management API, or build on the same tools your agents use.

import { GraphQLClient, gql } from 'graphql-request'
const query = gql`
query ListDocuments {
documents(includeDrafts: false) {
id
isPublished
metaDescription
metaTitle
publishedAt
slug
subtitle
title
}
}
`
const client = new GraphQLClient('https://content.pmkin.io/graphql', {
headers: {
Authorization: `Bearer ${process.env.PMKIN_API_KEY}`
}
})
const { documents } = await client.request(query)

Bring your team. Bring your agents.

Start for free, invite your teammates, and connect your first agent in a couple of minutes.