Add hover, upload, marked shots, and richer page walk - #3
Merged
Conversation
Agents can pierce shadow roots and iframes, heal stale refs, hover menus, assign /home/agent files to file inputs, and read page text from navigate and click. Marked screenshots overlay snapshot numbers without changing the live DOM. Co-authored-by: Cursor <cursoragent@cursor.com>
This was referenced Aug 26, 2026
Partial fills no longer post the form, and upload treats `..` as a path segment so names like report..final.pdf still work. Co-authored-by: Cursor <cursoragent@cursor.com>
…space. Page text now comes from the same settled page as the returned snapshot, and iframe controls are no longer hidden by unrelated top-level rectangles. Co-authored-by: Cursor <cursoragent@cursor.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
computer_hoverandcomputer_upload(path under/home/agent/, ≤5MB, viaGET /file), auto-activate a click-opened tab, and return the first 2000 chars of page text from navigate/click.marks=trueoverlays snapshot boxes on screenshots (Pillow 12.2.0). MCP, Drive tool schemas, and the developer prompt name the new verbs.Test plan
python tests/test_browse.pyandpython tests/test_navigate.pynode web/web-ui/test_serve.mjs/home/agentintoinput[type=file]computer_screenshot marks=truedraws numbered boxes; a bad PNG still returns the original image/home/agent, password-like inputs, and files over 5MBMade with Cursor