Solutions

Writers & creators: topics, drafts, publishing in one place

From topic backlog to interview transcripts, drafts, and a publishing calendar — everything lives in Kition. The AI agent curates topics, rewrites drafts in your voice, and adapts a single piece for multiple platforms. No more copy-paste across five tools.

One local vault for the whole content lifecycle.

Who it is for

Independent writers, newsletter authors, social-media teams. Shipping weekly, sensitive to both productivity and data ownership.

What hurts today

Notion for the topic board, Obsidian for drafting, Google Calendar for scheduling, ChatGPT for polishing — four tools, four tabs. The topic board and drafts are not really linked, the AI never sees your voice, and platform adaptation is manual.

How Kition helps

Kition consolidates the topic table, drafts, and publish calendar in one vault. The agent reads your back catalog and writes in your voice. It can pull "last week’s high-rated drafts" from the topic table. It can fan one piece out into Twitter / newsletter / LinkedIn versions in parallel.

Suggested workflow

Step 1

Topic backlog → table

Make a "topics" table with title, keywords, status, platform, deadline. The agent suggests what to write today based on trends and your history.

Step 2

Drafts → Markdown

One .md per topic. The agent uses apply_patch so every change has a visible diff.

Step 3

Publish calendar → Calendar view

The same table flipped into Calendar view becomes your editorial calendar.

Step 4

Platform adaptation → subagents

Need three versions of the same piece? Spawn three subagents in parallel — each handles one platform’s tone and length.

Outcomes

  • Whole pipeline lives in one tool
  • Drafts never leave your machine; platform accounts stay yours
  • Agent learns your style from your vault history
  • Cross-platform adaptation goes from manual to ⌘K

Ready when you are.

Kition is a local-first AI workspace. Markdown documents, structured tables, and an AI agent — running on your own machine, against the model provider you choose.