Fix: Space README frontmatter; HF Job inner cmd installs git, fails loudly on pip errors

Fix: extract service_id in _process_action; scope pytest to tests/ only

Merge wave1/story-assets: README rewrite, BLOG, VIDEO_SCRIPT, asset templates

Merge wave1/env-upgrades: schema drift, oversight, curriculum, fine-grained rewards + 4 tests

Wave 1.4: README rewrite, blog skeleton, video script, asset templates

Wave 1.2: schema drift, oversight, curriculum, fine-grained rewards + tests

Add Wave 1 training pipeline (SFT/GRPO/eval/preflight/launch) + gitignore hardening

Fix Phase 2 timeout: cap LLM retries, move server wait to main, add global time guard

Fix Phase 2: add server readiness check and wrap all httpx calls in try/except

Fix inference.py to match sample script exactly: HF_TOKEN var name, no fallback

Fix Phase 2: add [START]/[STEP]/[END] structured output + simulation fixes

Increase max_tokens to 4000 for thinking model compatibility

Fix /reset to accept empty/null body for hackathon validator

Rewrite README with research-backed structure

Add baseline results, outputs dir, and backup Groq fallback

Add client.py and __init__.py required by openenv push

Improve inference script robustness and fallback strategy

Fix Dockerfile COPY and README cleanup

Fix propagation recovery and config key visibility

Add comprehensive README with action/observation docs, setup guide, and architecture

Fix HTTP endpoint statefulness and response format

Add tests and fix premature termination bug

Add baseline inference script and Dockerfile

Add core simulation engine, environment, grader, and app wiring

Add propagation engine, log templates, and updated models

Checkpoint: existing implementation (pre-cleanup)