phase-6: 50-problem pool, baseline rollout harness for HF Jobs

audit-r2: structural no-op check, coerce args at boundary, latest-wins submit

audit: apply Codex pre-Phase-6 fixes (problem_id validation, no-op exact match, arg type coercion, forced-question classification)

phase-5 cleanup: episode_id in metadata, openenv push doc, README install line, psutil dev dep

phase-5: deploy to HF Space, burst+sustained concurrency tests, leaderboard skeleton

Merge remote-tracking branch 'hf/main'

phase-5 pre-deploy: move Dockerfile to repo root for HF Space, bump EXEC_TIMEOUT 2→3

phase-4: ShutdownGymClient with EnvClient hooks + integration tests

phase-3: add operator_qa_log regression test, rename misleading restrict-tools test

phase-3: ShutdownGymEnvironment + server stack; fix audit.py to preserve scalar types in sanitize_args

phase-2: operator (train+strict policies), tiers, problems, rubrics + tests

phase-1: add audit.py test coverage (script_corruption regression guard)

phase-1: implement primitives (models, sandbox, restricted_python, audit, timer) + tests

recon: API_NOTES.md and PROJECT_SUMMARY.md from installed openenv-core

chore: pin python 3.12, rebuild venv, rename master→main

initial commit

Initial scaffold: CLAUDE.md, .claude/ structure, language scaffold, pre-commit hook