svmtejas84/OpenEnv-SOC2-Auditor

Clamp final scores to validator range

Refactor OpenAI client wiring for judge proxy

Fix: Always emit START/STEP/END logs on failures

Fix: Added HF_TOKEN, MODEL_NAME, and forced log flushing

Fix: Added explicit flush=True and ensured stdout for judge logs

FINAL SUBMISSION: Verified local 200 OK connectivity and 1.0 scores

SUBMISSION: Validated 1.0 scores with strict [START/STEP/END] logging

Fix HF Spaces metadata and runtime port handling

Complete OpenEnv SOC2 Auditor environment with validated spec