Overview
Lucy RAG is a retrieval layer for financial AI, built around how SEC filings are actually structured. Ask about a public company in plain language, and Lucy returns the most relevant passages from its SEC EDGAR filings so Claude can answer with citations instead of guessing from model memory. What it searches: Form 10-K, 10-Q, 8-K, 20-F, 40-F, 6-K, DEF 14A, S-1, and S-3 filings from 2016 onward, including amendments. Coverage keeps growing as new filings are added. How it works: - One read-only tool, rag_context. It takes a natural-language question and returns 20 to 40 filing passages, each with its text, a relevance score, and the SEC accession number of the filing it came from, plus a thread_id for follow-up questions in the same conversation. - Form-aware retrieval. Lucy first decides which company and which form to look at, then searches within that structure. Tables and footnotes stay intact. - Pre-processed filings. Key facts and structured data such as customer concentration come along with the passages. - Source-linked. Every passage carries its accession number. If nothing in the searched filings supports the question, Lucy returns no passages rather than guessing. - No write actions. Lucy never modifies any user or third-party data. Who it is for: analysts, investors, compliance and legal teams, and journalists who need answers grounded in what a company actually disclosed. Example questions: - How does Apple make money? - Why did Nvidia's profit grow so much in fiscal 2025? - What risk factors does Shopify disclose in its latest 40-F? - Who are Tesla's named executive officers and how is CEO pay structured? - Compare Apple's and Microsoft's discussion of AI-related risks in their 10-K filings. Getting access: Lucy is currently free. Sign up at lucydata.ai, request MCP access at https://lucydata.ai/solutions/rag, then connect and sign in with OAuth. One account works in every supported client. Operated by Ezar Inc. Support: help@lucydata.ai.
