Drop a Zoom or Otter transcript into a directory the agent watches. It identifies decisions, action items, owners, and deadlines. Posts the summary to a Slack channel and writes the markdown to a docs folder. The structured output is the thing teams actually use a week later.
1modules:2 filesystem: {}3 workspace:4 config:5 render_mode: markdown6 sync_to_disk: true7 channels: { config: { slack: { bot_token: { credential: slack_bot } } } }89triggers:10 - id: new-transcript11 kind: file_watch12 path: ~/transcripts/13 target: { agent: notes, message: "{{event.path}}" }1415agents:16 - id: notes17 modules: [filesystem, workspace, {channels: [slack_post]}]18 brain: { model: claude-sonnet-4-6 }# 1. install runtime
curl -sSL https://digitorn.ai/install | sh
# 2. save the YAML above to ~/.digitorn/apps/my-meeting-notes/app.yaml
mkdir -p ~/.digitorn/apps/my-meeting-notes
# paste the YAML into app.yaml
# 3. deploy
digitorn deploy my-meeting-notesEngineering notes from the Digitorn team. No marketing, no launch announcements, no "10 prompts that will change your life". Just the things we write that we'd want to read.