feat(vision): add /find endpoint for template/text target detection #2

Open
opened 2026-04-05 20:16:08 +02:00 by luna · 0 comments
Collaborator

Summary

Add a helper endpoint that finds candidate UI targets (text/template/icon) and returns bounding boxes.

Why

Reduces guesswork for coordinates and makes interactions less brittle.

Scope

  • Add
  • Support matching by text (when OCR enabled) and by template image
  • Return array of matches sorted by confidence

Acceptance criteria

  • Can find a provided template on current screen
  • Can find text matches when OCR data is available
  • API docs include usage pattern with clickthrough actions
## Summary Add a helper endpoint that finds candidate UI targets (text/template/icon) and returns bounding boxes. ## Why Reduces guesswork for coordinates and makes interactions less brittle. ## Scope - Add - Support matching by text (when OCR enabled) and by template image - Return array of matches sorted by confidence ## Acceptance criteria - Can find a provided template on current screen - Can find text matches when OCR data is available - API docs include usage pattern with clickthrough actions
Sign in to join this conversation.
No Label
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: space/clickthrough#2