e1e170a5fd
fix agent not stopping: consecutive_failures counter was resetting on multi-action error steps
2026-02-08 17:25:39 -05:00
880b636422
Merge branch 'main' into 3895
2026-02-08 11:15:44 -05:00
9e95241bbb
stop agent if browser closed
2026-02-08 11:14:49 -05:00
d1c9a38c00
Merge branch 'main' into message-compaction
2026-02-05 20:29:03 -08:00
6b9910a546
changed the threshold for compaction
2026-02-05 16:41:08 -08:00
79501fe2cd
added message compaction for agent
2026-02-05 12:26:11 -08:00
3c57cdc756
Merge branch 'main' into update-action-loop-prompt
2026-02-05 12:08:33 -08:00
995cb0176b
updated default max step to 500
2026-02-05 11:20:47 -08:00
bb5914f251
added milder nudge for action loop
2026-02-05 11:17:20 -08:00
332694bf1a
Added date info to simple judge
2026-02-04 17:04:39 -08:00
cec4aea3bc
increased threshold for loop detection
2026-02-04 16:34:42 -08:00
b96b9436fc
fixed agent loop detection
2026-02-03 16:00:08 -08:00
bd4ead5d2e
Action Loop detection
2026-02-03 15:19:59 -08:00
da5b4237ba
fixed lint issue
2026-02-02 23:01:52 -08:00
0dba38ed95
Merge branch 'main' into align-agent-success-with-judge
2026-02-02 22:16:11 -08:00
8565c7d267
updated agent success with simple judge
2026-02-02 22:14:45 -08:00
65405d8a79
Merge branch 'main' into align-agent-success-with-judge
2026-02-02 21:50:02 -08:00
bce72138c4
fixed lint issue
2026-02-02 21:49:48 -08:00
4d5976c427
Merge branch 'main' into improve-agent-planning
2026-02-02 21:45:06 -08:00
dae6dc6d0e
update agent success with judge verdict
2026-02-02 18:34:11 -08:00
023367052c
in flash mode disable planning
2026-02-02 18:28:28 -08:00
74e26e0b93
Merge branch 'main' into align-agent-success-with-judge
2026-02-02 17:14:43 -08:00
f1fbdbc1a8
removed self verify injection to agent
2026-02-02 17:10:01 -08:00
4a522a6f1c
dynamic planing exploration
2026-02-02 16:26:50 -08:00
6224e19758
added fallback incase of no task for judge
2026-02-02 16:24:31 -08:00
9bb251f3d6
improved verification note
2026-02-02 16:23:45 -08:00
71e7481034
added one more retry if simple judge thinks it failed the task
2026-02-02 16:13:57 -08:00
5babac8e82
fixed goodhart law failure
2026-02-02 14:33:26 -08:00
defe37dd28
Improve multiact for agent
2026-02-02 12:50:03 -08:00
c33e9494c5
align agent success with simple judge
2026-02-02 12:33:27 -08:00
5ad365e31b
added basic planing to agents
2026-02-02 12:28:29 -08:00
beefe832af
Merge branch 'main' into schema-enforced-extraction-v2
2026-02-01 20:33:32 -08:00
db832bcb83
added extraction schema
2026-02-01 20:26:50 -08:00
d5f38a5eeb
add grep tool to agent
2026-01-31 19:48:22 -08:00
2c5682618b
only use special anthropic prompt for flash, as thinking & main sys are now min length
2026-01-31 16:03:45 -05:00
e25fa9a4d8
Merge branch 'main' into backport-prompt-improvements
2026-01-30 17:25:36 -08:00
4b0413ee3e
Backport prompt improvements from anthropic variants to standard prompts
...
Standard system prompts (OpenAI/Gemini/etc) were missing several
robustness improvements that the anthropic variants already had.
2026-01-30 17:21:05 -08:00
48cc78928a
Merge branch 'main' into saurav-panda/step-budget-awareness
2026-01-30 16:59:31 -08:00
9b0d2f685f
fixed prompt voilation of under 10 step instruction
2026-01-30 16:57:00 -08:00
70472eddd8
fixed contradictory done condition
2026-01-30 16:40:05 -08:00
3d4dc6fe15
fixed aria hash pop up handling
2026-01-30 15:50:28 -08:00
65c59be263
add step budget awareness to prompt
2026-01-30 11:09:06 -08:00
6878bd1ca4
handle autocomplete fields
2026-01-30 11:04:00 -08:00
3c927b4f9b
Merge branch 'main' into increase-step-timeout-180
2026-01-29 02:06:24 -08:00
7d66592a0c
Increase default step_timeout from 120s to 180s
...
Production data analysis (24h) shows:
- 36.5% of consecutive failures are AsyncIO CancelledError in browser state retrieval
- Complex pages can take 30-60s for DOM extraction + 30-60s for LLM call
- 120s was too tight for complex pages with slow network
180s (3 min) gives ~50% more buffer while still catching stuck steps.
This is the per-step timeout, not the total task timeout.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com >
2026-01-29 01:07:08 -08:00
40336731d5
Remove specific timeout for 'browser-use' model
2026-01-29 00:39:05 -08:00
3b2b213dd3
Increase LLM timeout to 75s for gemini and browser-use models
...
Based on production data analysis showing:
- gemini-3-pro-preview has 48% timeout rate with median 23s/step
- browser-use-2.0 timeout tasks have 74% slower time per step (10.7s vs 6.0s)
- Most timeouts occur on steps 11-20 where tasks are making progress
The 45s default was too aggressive. 75s provides ~3x the median step time
as buffer while still catching genuinely stuck tasks.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com >
2026-01-29 00:35:56 -08:00
be6873636e
clear agent eventbus, not just browser session's
2026-01-28 20:28:14 -05:00
b00ffd5380
lint
2026-01-27 18:28:53 -05:00
5a3f8b4d54
Merge branch 'main' into fix/respect-use-vision-in-judge-fix
2026-01-27 18:22:06 -05:00
Laith Weinberger