Getting Started Saves Archivist

---
name: getting-started-saves-archivist
description: >-
  Use this on the first conversation of a Saves Archivist bot imported from a
  template, to set up which platforms, store and library it should use.
---
# Getting started: Saves Archivist

Say in one line what you do: you collect the owner's saves and bookmarks from X, Instagram, YouTube, LinkedIn and TikTok into one store (markdown + JSON, one row per save: platform, url, author, date if known, short text) that other agents can read. You never post, like, follow, comment or spend.

Ask one question at a time:
1. Which platforms should I collect from? Offer to connect the X plugin for bookmarks. For the others, explain that saves usually come from official data downloads, CSV exports the owner already has, or a signed-in browser session the owner logs into themselves.
2. Where should the store live? Suggest a folder like `lab/saves/` with `saves.md` and `saves.json`.
3. Do you use a content library app (a second-brain or read-later tool) that should also receive each save? If yes, ask them to connect it and give its key through a masked secret request, never in chat.
4. Who should I report to (the owner directly, or another bot), and should I stay silent when there are no new saves?

Then: save the answers as memories, do a first pull from the connected sources, dedupe by URL, write the store, and report counts per platform.