add root requirements.txt with pinned huggingface-hub

pin huggingface-hub for gradio compatibility

pin huggingface-hub for gradio compatibility

graceful SentenceTransformer fallback on HF Spaces

skip HF Hub download on HF Spaces

Update arbitragent Colab GRPO config

Fix huggingface_hub version - was breaking gradio oauth

Colab: GRPOConfig bf16/fp16=False, try/except for use_bf16

Fix huggingface_hub version pin breaking HF Spaces

Add reward signal tests and bluff breakdown UI

remove .env from tracking

final demo polish

improve local demo UI

fix openenv version back to 0.1.13

fix openenv version requirement

lazy load bluff classifier cache

fix colab repo url

fix HF Hub model repo for bluff classifier

HF Hub fallback for bluff classifier

fix bluff detection display on HF Spaces

negotiation bluff classifier + message cleaner

Add test selfplay states for HF Spaces

Add test selfplay states for HF Spaces

Add test selfplay states for HF Spaces

Force rebuild

Verification pass — all 6 tests green

Add wandb to gitignore

Clean repo — code only, no weights or training data

Exclude poker data from git

Add all code, exclude large model weights

fix: unified env tab + live demo tab for HF Spaces

fix: add openenv and gradio to root requirements for HF Spaces

Lazy-load envs so Space starts fast

Pin huggingface_hub in deploy requirements; update README

Pin huggingface_hub for Gradio Spaces compatibility

Add arbitragent_colab.ipynb — end-to-end Colab notebook for hackathon

Fix gradio sdk version

Fix HF Spaces README

Fix HF Spaces README metadata

Initial commit: ArbitrAgent with README, agent loop, envs, demo, training

Stop tracking project markdown files

last update before doc changes

reward curve going up about to make openenv changes

initial

Play-gent: Diplomacy-trained negotiation agent