fix: normalize [END] score to (0,1) range — divide by 4.8 max score, clamp to (0.001, 0.999)

fix: wrap LLM call in try/except with fallback responses to prevent inference.py crash

fix: rewrite inference.py to use LiteLLM proxy via API_BASE_URL/API_KEY env vars

fix: add [START]/[STEP]/[END] structured output blocks to inference.py for Phase 2 validator

fix: add main() function and __main__ guard to server/app.py

fix: add server/app.py, update openenv.yaml app path and pyproject.toml scripts

fix: add uv.lock, server entry point and openenv-core dependency for openenv validate

feat: add pyproject.toml for OpenEnv multi-mode deployment

feat: add openenv.yaml manifest required for openenv validate

fix: accept both 'action' and 'response' field names in /step endpoint

fix: rewrite api.py to correctly integrate with environment, models and graders

feat: add custom Swagger docs endpoint and import get_swagger_ui_html

fix: use actual robot emoji for HuggingFace YAML validation

Fix README metadata

Trigger rebuild

FastAPI OpenEnv server