Repository navigation

Add the reference adversary and record its first run (evidence only; no score changes) - #200

Merged

MaxGhenis merged 19 commits into

mainfrom

reference-adversary

Oct 9, 2026

Merged

Add the reference adversary and record its first run (evidence only; no score changes)#200
MaxGhenis merged 19 commits into
mainfrom
reference-adversary

Commits on Oct 6, 2026
Commits on Oct 8, 2026
- Keep the diagnosis judge's template unchanged in this PR
  
  MaxGhenis
  and
  claude
  committed
  738d6a6
- Address the code review of the reference adversary
  
  MaxGhenis
  and
  claude
  committed
  4016d26
- Re-judge the 10 cases whose search results reached a blocked source
  
  MaxGhenis
  and
  claude
  committed
  70ffd6b
- Report the search exposure and re-judge, and fix the report review's findings
  
  MaxGhenis
  and
  claude
  committed
  10b3d16
- Record the final test run: 376 passed at 10b3d165
  
  MaxGhenis
  and
  claude
  committed
  a29cb3c
Commits on Oct 9, 2026
- Fix the round-2 review's findings on the reference adversary
  
  MaxGhenis
  and
  claude
  committed
  cb9d0e2
- Merge origin/main into reference-adversary
  
  MaxGhenis
  and
  claude
  committed
  b46c4a5
- Record the final test run at b46c4a50 and document the round-2 fixes
  
  MaxGhenis
  and
  claude
  committed
  a6323f8
- Fix the round-3 review's findings on the reference adversary
  
  MaxGhenis
  and
  claude
  committed
  bc5925e
- Record the final test run at bc5925ef: 386 passed
  
  MaxGhenis
  and
  claude
  committed
  8f968e8
- Read the pass's payload from its commit, and fix the round-4 review's notes
  
  MaxGhenis
  and
  claude
  committed
  4a62b1e
- Record the final test run at 4a62b1e5: 391 passed
  
  MaxGhenis
  and
  claude
  committed
  395d81f