Start here if you're new to FLS:
INSTALL.md— installation options and verification steps.USAGE.md— the main workflow (--delta, integrity checks, summarization).docs/API_KEYS.md— Semantic Scholar key + Unpaywall email setup (and.env.local).docs/RAG.md— local/offline RAG (index + query), plus OCR option for scanned PDFs.docs/UNPAYWALL.md— Unpaywall endpoint details and usage notes.docs/CUSTOMIZATION.md— tuning keywords/categories/research context.docs/AGENT_INTEGRATION.md— agent session protocol (model-agnostic) and templates.templates/AGENT.template.md— copy/paste protocol for any agent instructions file (model-agnostic).
Typical loop:
fls.py(optional convenience runner; confirms config then runs scan + index)literature_scan.py --delta(fetch + download PDFs)rag_index.py(build/update local index)rag_query.py(retrieve evidence/snippets for synthesis)orchestrate_summarization.py(agents read PDFs and store structured notes)