Commit Graph

320 Commits

Author SHA1 Message Date
Marc Kelechava
4401216346 add step count to webhooks and get run payload (#4410) 2026-01-07 11:41:57 -08:00
Celal Zamanoglu
058a9178aa link actions to their screenshots - backend (#4404) 2026-01-07 02:12:22 +03:00
LawyZheng
f592ee1874 bulk post action artifacts (#4327) 2025-12-19 02:16:44 +08:00
Stanislav Novosad
1eca20b78a Batch LLM artifacts creation (#4322) 2025-12-17 20:15:26 -07:00
LawyZheng
c9c66398c4 Revert "File Download renaming reliability - customer bug fix" (#4311) 2025-12-17 14:54:13 +08:00
Marc Kelechava
7557a130a3 File Download renaming reliability - customer bug fix (#4308) 2025-12-16 15:20:31 -08:00
LawyZheng
ce717146f3 reenbale the download action (#4299) 2025-12-15 14:30:32 +08:00
LawyZheng
2de27637db fix hover action (#4245) 2025-12-10 02:39:17 +08:00
Mohamed Khalil
f49b07f30d feat: add hover action support (#3994)
Co-authored-by: LawyZheng <lawyzheng1106@gmail.com>
2025-12-09 23:27:26 +08:00
pedrohsdb
40c8f39045 feat: remove ENABLE_PARALLEL_USER_GOAL_CHECK experiment in favor of treatment (#4235) 2025-12-08 18:12:53 -08:00
pedrohsdb
fadfe0848c fix speculative artifact persistence (#4211) 2025-12-05 08:39:36 -08:00
LawyZheng
0cf52486de fix totp code bug (#4210) 2025-12-05 21:57:14 +08:00
Stanislav Novosad
f754272f9c Extract BrowserState.scrape_website (#4184) 2025-12-03 15:08:32 -07:00
pedrohsdb
ce01f2cb35 fix: prevent Vertex cache contamination across different prompt templates (#4183) 2025-12-03 11:13:27 -08:00
Stanislav Novosad
19d2deb859 Split browser_state/manager protocol and implementation (#4165) 2025-12-02 11:08:38 -07:00
LawyZheng
994461f69c fix recording for browser session run (#4161) 2025-12-02 14:29:00 +08:00
pedrohsdb
3f11d44762 Pedro/fix vertex cache leak (#4135) 2025-11-29 07:39:05 -06:00
Stanislav Novosad
a820bb6daa Add failure_reason to "Task duration metrics" (#4122) 2025-11-27 16:13:58 -07:00
pedrohsdb
4fc8838730 remove skip screenshot annotations experiment (#4111) 2025-11-26 14:43:58 -08:00
LawyZheng
8d09d9822a clean up fullpage screenshot exp (#4102) 2025-11-26 14:55:02 +08:00
LawyZheng
e692ae8944 file download block should not trigger parallel check (#4100) 2025-11-26 14:23:41 +08:00
LawyZheng
d3fe5e1b02 wait for animation ends before taking post screenshot (#4098) 2025-11-26 13:55:46 +08:00
LawyZheng
21d5cd0f18 rollout termination aware exp for everything (#4084) 2025-11-25 03:41:50 +08:00
LawyZheng
7dd8e1e4e0 Revert "scope termination-aware verification to file download fallback" (#4083) 2025-11-25 03:02:09 +08:00
Shuchang Zheng
2608c02f7a lower default page loading time from 90 seconds to 60 seconds (#4076) 2025-11-22 21:07:34 -08:00
pedrohsdb
c10016c8bc Respect disable goal check in parallel flow (#4021) 2025-11-21 15:07:50 -08:00
pedrohsdb
db68d8a60c scope termination-aware verification to file download fallback (#4043) 2025-11-19 17:34:08 -08:00
Stanislav Novosad
0efae234ab Initialize app at runtime instead of import time (#4024) 2025-11-18 17:56:58 -07:00
pedrohsdb
c561885bdd fix: ensure parallel verification runs data extraction (#4014) 2025-11-18 12:17:29 -08:00
LawyZheng
9b97061b6d fix log typo (#4016) 2025-11-18 14:44:09 +08:00
Shuchang Zheng
d118eb5d4e remove cache actions (#4015) 2025-11-17 21:06:51 -08:00
LawyZheng
abcdf6a033 support download by select action (#4009) 2025-11-17 14:46:32 +08:00
Shuchang Zheng
25e375f78f execute_task_webhook uses the latest non canceled step (#4007) 2025-11-16 15:01:40 -08:00
LawyZheng
9814f9803a fix error reason when page is no data (#3998) 2025-11-14 12:38:11 +08:00
pedrohsdb
b7e28b075c parallelize goal check within task (#3997) 2025-11-13 17:18:32 -08:00
LawyZheng
40cc6c7b47 support non url task block (#3983) 2025-11-13 14:28:45 +08:00
pedrohsdb
d88ca1ca27 Pedro/vertex cache minimal fix (#3981) 2025-11-12 10:40:52 -08:00
pedrohsdb
ca958da6be Add termination-aware complete verification experiment (SKY-6884) (#3948) 2025-11-07 18:53:51 -08:00
pedrohsdb
d8631151ba Speed optimizations: Economy element tree and TOTP context parsing skip (#3936) 2025-11-06 21:56:52 -08:00
pedrohsdb
44528cbd38 Pedro/fix explicit caching vertex api (#3933) 2025-11-06 14:47:58 -08:00
pedrohsdb
d2f4e27940 Add feature flag to skip screenshot annotations (#3932) 2025-11-06 12:46:32 -08:00
Marc Kelechava
3db5ec6cd7 [SKY-6974] Browser Profiles [2/3] Marc/backend browser session profiles (#3923) 2025-11-06 01:24:39 -08:00
LawyZheng
7ff809e50b refactor webhook signature (#3889) 2025-11-04 11:29:14 +08:00
pedrohsdb
06bb9efb4a parallel check user goal xp (#3873) 2025-10-31 12:19:50 -07:00
pedrohsdb
0e0ae81693 Improve LLM error message when LLM is down (#3874) 2025-10-31 11:41:07 -07:00
pedrohsdb
76de33edbd removing laminar (#3858) 2025-10-29 21:42:27 -07:00
pedrohsdb
b89b882d6e set up xp for using cheaper model for verication result (#3853) 2025-10-29 15:11:40 -07:00
Shuchang Zheng
b1baacf138 fix reload action (#3811) 2025-10-24 21:06:13 +00:00
Shuchang Zheng
d55b9637c4 set context.step_id and context.task_id at the beginning of execute_step and unset at the end + auto log step_id & task_id (#3803) 2025-10-23 16:32:28 -07:00
LawyZheng
87625d4c0f support new tab magic link logic (#3797) 2025-10-23 14:38:03 +08:00