Skip to content

Use Cases

Z.E.N. isn't a chat tool. It's the layer underneath whatever AI surface you already use, running the meals you've made enough times to write down. Three patterns show up over and over.

The morning brief

The pattern most people start with. Every weekday at 8 AM, Z.E.N. pulls the messages you actually need to read out of Slack, the docs that changed overnight out of Notion, and combines them into one short brief. Lands in your Telegram before you open your laptop.

yaml
name: morning-brief
schedule:
  cron: "0 8 * * 1-5"
  timezone: "America/New_York"

nodes:
  - id: pull-slack
    type: prompt
    prompt: |
      Summarize unread messages from #team and #leadership since 6pm yesterday.
      Quote anyone who tagged me directly.

  - id: pull-notion
    type: prompt
    prompt: |
      Find pages edited in the last 24 hours under Strategy and Product.
      Return titles plus a one-line summary each.

  - id: combine
    type: prompt
    depends_on: [pull-slack, pull-notion]
    prompt: |
      Combine the Slack summary and the Notion edits into a single brief.
      Headline at top, sections below. Post the result to Telegram.

Who runs this: founders, ops leads, anyone who manages on Slack and writes in Notion. The pattern scales: swap the connectors, swap the format, run it for a different team at a different hour.

The deep research sprint

For when a strategic question lands and you need a real answer, not a vibes-based one. Z.E.N. runs three rounds of research (mainstream sources, contrarian takes, expert-only pulls), then drafts a one-page memo with a headline, three sections, and a recommendation. Five minutes from prompt to memo.

yaml
name: research-sprint
description: Three-round research with a final memo

nodes:
  - id: round-one
    type: prompt
    prompt: |
      Search the web for the most authoritative recent sources on "$TOPIC".
      Return five sources with one-line summaries.

  - id: round-two
    type: prompt
    depends_on: [round-one]
    prompt: |
      Find three contrarian or critical takes on the same topic.
      Sources only experts would cite. Same format as round one.

  - id: round-three
    type: prompt
    depends_on: [round-one]
    prompt: |
      Find three niche or industry-specific sources that the mainstream
      results missed. Same format.

  - id: memo
    type: prompt
    depends_on: [round-one, round-two, round-three]
    prompt: |
      Combine all three rounds into a one-page memo. Headline at top.
      Three sections: what the mainstream view is, where it's contested,
      what to actually do. Bottom-line recommendation in two sentences.

Who runs this: researchers, analysts, strategists, anyone who gets handed open-ended questions and has to come back with a defensible take by end-of-week.

The content cadence

A pipeline for people who ship writing. Topic in. Z.E.N. drafts an outline, expands it into a full piece, runs an editor pass against a style guide you provide, and (with approval) schedules it to wherever your content lives.

yaml
name: content-pipeline
description: Outline, draft, edit, schedule

nodes:
  - id: outline
    type: prompt
    prompt: |
      Outline a post about "$TOPIC". Headline, four sections, a call-to-action.
      Aim for a reader who is sophisticated but not technical.

  - id: draft
    type: prompt
    depends_on: [outline]
    prompt: |
      Expand the outline into a full draft. 800 words. Same audience.

  - id: edit
    type: prompt
    depends_on: [draft]
    template: zain-voice-edit
    prompt: |
      Edit the draft against the style guide. Cut anything that sounds AI.

  - id: approve
    type: approval
    depends_on: [edit]
    prompt: |
      Final draft. Approve to schedule, reject to send it back for another round.

  - id: schedule
    type: prompt
    depends_on: [approve]
    prompt: |
      Schedule the approved post to the content calendar in Notion.

Who runs this: copywriters, content marketers, anyone with a publishing cadence to hit and a style guide they're tired of explaining.

What these three have in common

Each pattern is something you could do by hand with Claude or ChatGPT. Each one stops working the moment you need it again next Tuesday. Z.E.N. is the layer that turns a one-off improvisation into something that runs while you sleep.

If one of these is close to a routine you already do by hand, copy the YAML, change the names, and fire it. The full vocabulary is in Concepts.

AI that follows a recipe, not a conversation.