fix: self-contained grader + remove FACTORY_TASK from Dockerfile

fix: lazy openenv.core imports so FactoryEnv works on all Python envs

fix: graders use FactoryEnv, pure-Python fallback if unavailable

fix: grader.py pure stdlib, no external deps, works on Python 3.11+

fix: standalone inference.py with raw websockets, correct obs/state parsing

fix: reconnect per task to avoid closed WebSocket on second episode

fix: zero-arg self-contained graders, inference runs all 3 tasks

fix: add spec_version:1 to openenv.yaml, add standalone root grader.py with score_easy/medium/hard

fix: rename API_KEY to HF_TOKEN, remove fallbacks, fix action spaces in STEP log

fix: add graders to openenv.yaml and clamp scores to (0, 1) exclusive

Fix local openenv validate: add server/app.py, project.scripts, uv.lock, rename server.py to main.py

Fix inference.py: rename IMAGE_NAME to LOCAL_IMAGE_NAME, move debug logs after [START]

Fix task selection: reset() accepts task param, all 3 tasks verified

Rebuild inference + client to match OpenEnv WebSocket spec

Serve Gradio UI + OpenEnv API from single Docker Space

Fix HF Space blank screen and state endpoint bug

Add HF Space metadata and openenv tag

OpenEnv Submission

update inference

Initial commit

first commit