A hands‑on RAG experimentation lab. Largely configurable with debug insights. Classification‑driven corpus construction, filter chains, document loading, chat interaction, Open WebUI integration. Experimental by design and not production‑ready.

Given a bug and a token budget, how should you pick which code to show the LLM? This repo contains a benchmark that evaluates different context assembly methods on that question, using controlled single-line mutations across two Python repositories. Still in development, will expand to more methods, models and repositories.