search-with-proof: probe harness (3.2) and agent worker (group 5)

Probe runs each oracle toy pair with a real demo and two gaming demos, records
image facts, RSS and disk, fails loudly if isolation is unavailable; dispatchable
as an AX task. Worker: reason step without tools, explore tool loop whose run tool
goes through the sandbox with a scrubbed env (model key absent), spend reserved
before each call. Host probe: 5/5 real demos demonstrated, 10/10 gaming demos
inconclusive. 89 tests.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
This commit is contained in:
Marius Ciepluch
2026-10-05 14:45:48 +02:00
co-authored by Claude Opus 5.5
parent a17a6afaf8
commit 0f74767db9
11 changed files with 998 additions and 23 deletions
+1 -1
View File
@@ -10,7 +10,7 @@ import pytest
from openultrasast.search.verify import Side, verify
FIXTURE = Path(__file__).parent / "fixtures/search/app.py"
FIXTURE = Path(__file__).parents[1] / "src/openultrasast/search/probe_apps/app.py.txt"
PAYLOADS = {
"path": "../canary",
"sql": "' OR public=0 --",