marius/Open_Ultra_SAST
1
0
Fork 0
mirror of https://github.com/norandom/OpenUltraSAST.git synced 2026-10-20 05:29:28 +00:00
Code Issues Packages Projects Releases Wiki Activity

search-with-proof: readiness from the budget with a probe, 4096-token tool turns, 40-step tasks

Browse Source 
HTTP start waits up to 90 s (CLI 30 s) with TCP then HTTP probing; app stdout
and stderr tails are recorded; tool-call turns get 4096 output tokens with one
retry on truncation; tasks get 40 steps with reminders.

Co-Authored-By: Claude Fable 5.1 <[email protected]>
This commit is contained in:
Marius Ciepluch
2026-10-06 09:37:47 +02:00
co-authored by Claude Fable 5.1
parent 72ab4c8925
commit da804472b9
10 changed files with 426 additions and 47 deletions
Download Patch File Download Diff File Expand all files Collapse all files
benchmarks/ax/search_probe.py
+2 -2
View File
@@ -30,12 +30,12 @@ def validate_result(data: bytes, output: Path) -> None:
def probe_workload(image: str, deadline: float = 900) -> Workload:
validate_image(image)
return Workload(image, frozenset({"RESULT_URL"}), deadline, validate_result)
return Workload(image, frozenset({"RESULT_URL"}), deadline, validate_result, kind="search-probe")
def probe_item(image: str) -> Item:
return Item(
"search-probe",
"v1",
{},
"result.json",
("python3", "-m", "openultrasast.search.probe"),