AkashKaralingannavar/sequential-medical-triage-ai

fix: add openenv-core>=0.2.0 dependency for validator compliance

fix: resolve syntax error and modularize grader into grader.py for OpenEnv compliance

fix: resolve syntax error and modularize grader into grader.py for OpenEnv compliance

fix: restore UI metadata (true_urgency/true_action) to step info dict

fix: absolute compliance overhaul with top-level grader and ultra-safe reward range [0.25, 0.75]

fix: total compliance overhaul with explicit graders, redundant reward models, and robust env-var support

fix: explicit task kwargs for grader discovery and robust env var support

fix: enforce single-task compliance and high-precision scoring for Phase 2

fix: total sum of rewards strictly between 0 and 1 for all difficulties

fix: clamp grading scores strictly between 0 and 1 for Phase 2 compliance

chore: structure repo as package with uv.lock for compliance

fix: add pyproject.toml for multi-mode deployment compliance

chore: sync inference.py manual edits and ensure compliance

refactor: make inference.py robust with try-finally and explicit env.close()

fix: enforce HF_TOKEN as mandatory and fix [END] line format for OpenEnv compliance

Initialize OpenEnv AI Medical Triage Assistant