Ground a screen design in real product references

Chooses whether the reference is a screen, a journey or a page section, searches twice (the exact feature, then another domain that solved the same problem), reads anatomy from the images, and cites each reference. For a designer or builder planning a screen.

What it asks for

  • about (Optional, words): What this run is for, in the person's words.
  • known (Optional, words): What is already known, decided or done, so the run starts from it instead of asking again.

The steps your AI walks

  1. Choose whether the reference is a screen, a journey or a page section
  2. Search twice: the exact feature, then a different domain that solved the same problem
  3. Extract anatomy from the images themselves, not from titles or metadata
  4. Cite every reference so the next person reproduces the same set, then record

Use it from your AI

Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this workflow by what it does and runs it.

Say to your AI: “Use Trillion to chooses whether the reference is a screen, a journey or a page section, searches twice (the exact feature, then another domain that solved the same problem), reads anatomy from the images, and cites each reference. For a designer or builder planning a screen”