What problem does it solve?
Filling out web forms in an automated browser often fails on non-text inputs like dropdowns, checkboxes, radio buttons, file uploads, and contenteditable fields, and submissions can silently fail without verification. This Skill provides a structured workflow for handling these complex form scenarios reliably.
Core Features & Use Cases
- Ref-Based Form Interaction: Use
select_option, set_checked, upload_file, type, type_submit, and submit actions against element refs captured via browser:content and browser:detail, avoiding fragile coordinate clicks.
- Visual Verification: Combine
browser:screenshot with vision_load to inspect layout, validation states, captcha-like UI, or canvas content that DOM inspection cannot reveal.
- Post-Submission Validation: Confirm results with
browser:content, browser:state, or a fresh screenshot after submission.
- Use Case: Imagine automating a multi-step signup form with a dropdown country selector, terms-of-service checkbox, and avatar upload. This Skill guides the agent to capture refs, fill each field with the correct action, verify the file path exists before upload, and confirm successful submission visually.
Quick Start
Fill out the signup form on the current page, including the dropdown, checkbox, and profile photo upload, then verify the submission succeeded.
