SHUBHAMOS/meta-pytorch-hackathon

fix(eval): default --task to 'all' so validator sees 5 tasks on bare python inference.py call (7th April)

fix(eval): update structured stdout blocks to exact format [START]/[STEP]/[END] and add flush=True

fix(eval): update env variables and add START/STEP/END structured logs (7th April)

fix: resolve NameError for primary_client in inference.py

Compliance: Rename entry point to main() and fix script registration

Fix boot-time path for inference script in server app

OpenEnv Compliance Fix: Package structure, relative imports, and uv.lock generation

fix(multi-mode): correct dependency name to openenv-core 0.2.x

fix(multi-mode): restructure repo for OpenEnv validation compliance

fix(prod): rename Nvidia, optimize port binding, and add pyproject.toml

feat: satisfy OpenEnv multi-mode with pyproject.toml and optimize startup speed

fix: resolve 422 Unprocessable Entity by making /reset body optional

docs: finalize stunning README and Judges Evaluation Center

fix: resolve EmailCategory validation crash and harden Action model

fix: Gradio Table attribute error and AI credit depletion logic

feat: health check system & e2e startup validation (removed secrets)

feat: unified Gradio frontend + FastAPI backend (interactive benchmark)

fix: extreme Dockerfile simplification (diagnostic build)

fix: docker build optimization (pinned numpy + build deps)

fix: add root health endpoint for HF Spaces

fix: docker cmd port fallback

fix: hf space metadata and clean repo

clean: remove pycache from repo

fix: hf space config error (port + startup)

resolve merge

SHUBHAMOS

initial commit

feat: robust LLM loop with fallback action parsing and backoff retry

docs: initialize v1.1 milestone (Integration & Hardening)

feat: complete SHUBHAMOS codebase

chore: complete v1.0 milestone

docs(phase-06): complete phase execution

test(phase-06): verify dashboard

docs(phase-06): plan dashboard

docs(phase-05): complete phase execution

test(phase-05): verify dockerfile and docs

docs(phase-05): plan dockerfile and readme

docs(phase-04): complete phase execution

test(phase-04): verify inference script

docs(phase-04): plan inference script

docs(phase-03): complete phase execution

test(phase-03): verify tasks and graders

docs(phase-03): plan tasks and graders

docs(phase-02): complete phase execution

test(phase-02): verify existing action space and reward engine

docs(phase-02): plan action space and reward engine

docs(phase-01): complete phase execution

feat(phase-01): bulk implement core environment models, engine, and spec

docs(01): capture phase context (auto mode)

docs: create roadmap (6 phases)