Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Default to Groq llama, NVIDIA 9B in registry as featured option

Fix model switcher: ACTIVE_LM + dspy.context

Tighten locked-state prompt to reduce rambling on small model

finals

Move portrait under contact in right column for balance

Update README.md

Update README.md

Update README.md

Update README.md

Add NVIDIA Nemotron model option, default to Nemotron

Update README.md

Track PNGs with Git LFS

Add portraits, reorder layout, center portrait

Add README and requirements

final copy done

Apply gitignore, add app.py, stop tracking db and env

feat: clue similarity, linked list progression, multi-NPC conversation loop complete

fixed retriever

added NPCNode, SQLite memory, linked list, clue similarity, best-of-N pipeline

fixed features, stage 1 complete

added config and main, improved judge

feat: constraint checker working, flag scanner + LLM judge returning PASS/FAIL

feat: full pipeline working, FAISS retr done, remove duplicate hits

first commit

Update README.md

initial commit