Commit Graph

939 Commits

Author SHA1 Message Date
Cursor Agent efc0f83063 Fix: Auto-configure screenshot size for browser-use models
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
2025-11-17 01:32:56 +00:00
Cursor Agent 487bd19e80 Refactor: Lower default max_actions_per_step to 4
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
2025-11-15 06:31:10 +00:00
mertunsall 7dc5f5bf24 we call the judge before done callback to enable in the cloud 2025-11-14 17:25:53 -08:00
Magnus Müller c29e358b9c Merge branch 'main' into kalil0321/demo-mode 2025-11-13 22:37:38 -08:00
Magnus Müller d454a67091 Merge branch 'main' into fix-upgrade-command-syntax 2025-11-13 22:14:39 -08:00
LarsenCundric deea28b4bb InterruptedError is due to user interrupting, so should be warning 2025-11-13 21:46:52 -08:00
kalil0321 0988199776 Merge branch 'main' into kalil0321/demo-mode 2025-11-13 15:34:49 +01:00
Balázs Thomay dc9dbbeddd Fix upgrade command to use == instead of @ for uv compatibility 2025-11-13 12:17:56 +01:00
Aitor 585141afd7 Merge branch 'main' into feat/improve-navigation 2025-11-13 00:41:51 +01:00
mertunsall dfbba0e062 add ground truth to judge 2025-11-12 11:16:40 -08:00
Cursor Agent 8f1b73335c Add 'polynomial' to excluded file extensions
Co-authored-by: mamagnus00 <mamagnus00@gmail.com>
2025-11-12 16:57:45 +00:00
Aitor 461d8592b1 Merge branch 'main' into feat/improve-navigation 2025-11-12 04:32:27 +01:00
mertunsall 2d5de65979 for some reason cant do self here 2025-11-11 16:18:32 -08:00
mertunsall ce9be49b32 change to self.logger 2025-11-11 16:13:33 -08:00
Cursor Agent cd4f96361a Refactor: Improve failure count message formatting
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
2025-11-12 00:06:52 +00:00
Mert Unsal ed23457f78 Update browser_use/agent/service.py 2025-11-11 16:03:02 -08:00
Cursor Agent 32adb3d095 Adjust logging level for agent failures
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
2025-11-12 00:01:04 +00:00
mertunsall a5b97a6cf4 this is not an error, just warning 2025-11-11 15:31:58 -08:00
mertunsall 018de804f7 fix comment 2025-11-10 21:30:39 -08:00
reformedot bf350bc4b0 refactor: update agent focus handling to use target ID consistently
- Replaced instances of `agent_focus` with `agent_focus_target_id` across various components to ensure consistent target management.
- Enhanced error handling and logging in `Watchdog` classes to reflect the new target ID approach.
- Improved session retrieval methods to streamline interactions with the `SessionManager`, ensuring better performance and reliability.
2025-11-10 14:29:50 -08:00
kalil0321 3f64178894 Implement demo mode for in-browser logging 2025-11-08 21:39:57 +01:00
mertunsall ed5128c302 sonnet has 1400x850 default screenshot size 2025-11-08 10:45:59 -08:00
mertunsall 94aeb5d385 add system prompt special to anthropic 2025-11-08 09:56:15 -08:00
mertunsall d06d8fa8a2 image resizing the make sonnet 4.5 work 2025-11-07 22:20:14 -08:00
mertunsall 8dc9a883d2 no need for screenshot tool when use_vision is True 2025-11-07 20:05:02 -08:00
mertunsall d3a88ce351 vision defaults to True 2025-11-06 16:29:09 -08:00
mertunsall 67d2f12665 default to PNG everywhere 2025-11-06 14:51:52 -08:00
Mert Unsal 8973e98ee0 Update browser_use/agent/service.py 2025-11-06 13:53:31 -08:00
Cursor Agent 714232bddd Add impossible task and captcha detection logging
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
2025-11-06 21:52:11 +00:00
Mert Unsal 2a9eae7ec5 Update browser_use/agent/service.py 2025-11-06 13:41:06 -08:00
Cursor Agent cb814da33e Add impossible_task and reached_captcha to JudgementResult
Co-authored-by: mailmertunsal <mailmertunsal@gmail.com>
2025-11-06 21:38:18 +00:00
mertunsall 1bb380734e improve judge printing 2025-11-03 17:06:59 -08:00
Cursor Agent 6f392b6775 Improve judge result console output to reduce noise
Only display judge results when:
- Self-reported success and judge verdict disagree
- Agent reports failure (shows full judge result)

Changes:
- Hide judge output when both self-reported success and judge agree task succeeded
- Show warning when agent reports success but judge thinks it failed
- Keep existing behavior for all other cases (show full judge result)

This reduces console clutter for successful tasks while highlighting discrepancies.
2025-11-04 00:09:42 +00:00
mertunsall 29ec2cacb6 improve judge logging to laminar and separate code 2025-11-03 13:04:34 -08:00
Alezander9 87e54e9e8e make use_judge on by default 2025-11-01 15:39:50 -07:00
Alezander9 bf15d159da address cubic comments and clean up new example 2025-11-01 15:37:38 -07:00
Alezander9 4396b416da pass along a new param request_type when using ChatBrowserUse to flag special handling for judge 2025-11-01 15:25:24 -07:00
Alezander9 4965e0d94c add library judge into telemetry event 2025-10-31 20:34:57 -07:00
Alezander9 10ec99dd26 have library judge show up in logs 2025-10-31 20:30:03 -07:00
Alezander9 15911c8c72 fix bugs in library judge and make it off by default 2025-10-31 20:20:36 -07:00
Alezander9 0f1946cb54 implement simple judge directly into the library on done action 2025-10-31 20:01:07 -07:00
Magnus Müller 78beda65fb Revert "Set vision to true by default"
This reverts commit 62ad269975.
2025-10-30 00:44:53 -07:00
Magnus Müller 62ad269975 Set vision to true by default 2025-10-29 22:46:46 -07:00
Magnus Müller 309c7dd8f5 Final error logging for not successful 2025-10-29 20:15:44 -07:00
Magnus Müller fafc274dde Fix log msg 2025-10-24 14:14:22 -07:00
Magnus Müller 258456736c Error at the end 2025-10-24 13:36:51 -07:00
Magnus Müller c13075d428 Msg if failed 2025-10-24 13:29:34 -07:00
Magnus Müller 0ff12c78c1 Merge branch 'main' into cursor/clean-url-endings-d286 2025-10-23 18:09:24 -07:00
Magnus Müller 3240b13501 reset state stopped after new task 2025-10-23 16:14:00 -07:00
mertunsall 0ac4ebff9c Fix laminar bug 2025-10-23 17:44:45 +02:00