substack-save_draft
Save a post as a draft on the person's own Substack publication, where they read and publish it themselves; nothing is published or emailed
What it asks for
- publication_host (Needed, words): The publication's host name, like name.substack.com; it must be the publication this account connected
- title (Needed, words): The post's title
- subtitle (Optional, words): The line under the title
- body (Needed, words): The post's words as Substack's post document, written as JSON text: {"type":"doc","content":[...]} where each block is a paragraph ({"type":"paragraph","content":[{"type":"text","text":"..."}]}), a heading ({"type":"heading","attrs":{"level":2},"content":[...]}), a bullet_list or ordered_list of list_item blocks each holding a paragraph, or a blockquote; a text may carry marks [{"type":"strong"}], [{"type":"em"}] or [{"type":"link","attrs":{"href":"https://..."}}]
- bylines (Needed, list): The draft's authors; Substack refuses a draft without one: [{"id": read_profile's id, "is_guest": false}]
- audience (Optional, one of): Who may read it once published; everyone when left out
Use it from your AI
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this tool by what it does and runs it.
Say to your AI: “Use Trillion to save a post as a draft on the person's own Substack publication, where they read and publish it themselves; nothing is published or emailed”
What your AI calls: run_tool {"tool_id":"tool:substack-save_draft"}