Skip to main content
Use capture tools to give Hercules precise visual context about what you want to change. Visual context eliminates ambiguity. Instead of describing where a button is, just show it. Hercules has 4 Capture Tools:
  • Screenshot: Take a screenshot of part of your app to reference specific UI
  • Annotate: Add comments, boxes, arrows, and notes to a screenshot of your app
  • Screen record: Record a video showing a bug or desired interaction
  • Select element: Point the Agent to a specific code component
Capture tools menu showing screenshot, annotate, screen record, and select element options

​
Screenshot

Screenshots allows you to the Agent visual context so it knows exactly what you’re referring to.

​
Annotate

Markup a screenshot of your app with comments, boxes, arrows, and drawings. Annotations make it extremely clear what changes you want to make and are exceptionally useful for quick UI and copy changes.
Annotating a screenshot with boxes and notes

​
Screen Record (coming soon)

Show a bug or desired interaction as a video. Screen recordings help the Agent understand timing, animations, and multi-step flows.

​
Select Element

Point the Agent at the exact HTML element (“DOM component”). This is useful when you want to change a specific element without ambiguity.

​
FAQ

Yes, you can use your own. However, Hercules compresses its screenshots so it reduces credit consumption.
Screenshot captures an image of your app. Select element identifies a specific HTML element / DOM component in the code.