Pradeerock/self-healing-rag

Include explicit score in inference logs

Clamp grader scores to exclusive bounds

Add score aliases to task metadata

Merge branch 'main' of hf.co:spaces/Pradeerock/self-healing-rag

Harden OpenEnv submission for resubmission

Merge branch 'main' into space-sync

Add explicit score to inference end logs

Merge branch 'main' into space-sync

Make HTTP task validation stateful

Merge branch 'main' into space-sync

Add explicit task graders and scoring

Merge branch 'main' into space-sync

Add root health response for Space

Merge Hugging Face Space history

Fix Phase 2 score range validation

Fix: clean inference and environment files

Fix: clean environment.py

Fix: clean inference and environment files

Fix: clean environment.py

Fix: clean inference.py no rag_env import

Fix: clamp rewards strictly between 0 and 1

Fix: clamp all rewards strictly between 0 and 1

Fix: all rewards strictly between 0 and 1 (not 0.0 or 1.0)

Fix: rewards strictly between 0 and 1

Fix: inference.py no rag_env import

Fix: no rag_env import use direct HTTP

Fix: inference.py uses direct HTTP not rag_env module

Fix: remove rag_env import, use direct HTTP

Fix: fully stateless verify

Fix: fully stateless verify

Fix: remove database_fixed gate from verify

Fix: single worker in Dockerfile

Fix: single worker in Dockerfile CMD

Fix: single worker for state persistence

Fix: force single worker for state persistence

Fix verify to not depend on database_fixed flag

Fix verify state check

Fix audit verify

Fix audit verify routing

Fix: database_fixed flag in audit verify

Add Task 4: Expert cross-topic database audit

Add Task 4 expert audit

Add task 4 expert audit

Self-Healing RAG Environment - OpenEnv Hackathon

Fix: set done in observation

Fix: add done field to RAGObservation

Fix: Pydantic models for HTTP serialization

Fix reset

Fix reset kwargs

Upload server/environment.py with huggingface_hub