The "book as an MCP server" framing is the part that interests me. I run a small MCP server myself (semantic search over government documents), and the thing I keep noticing is that the interface changes what people ask — nobody asks a PDF "what should I do in my situation," but they'll ask a tool that. Shipping specialized tools (draft_decision_doc etc.) over one generic ask_the_book tool also seems like the right call — clearer intent per call, less context spent on tool descriptions. I noticed you went with keyword search over a summary corpus plus citations rather than embedding the full text — curious whether that was a licensing decision (keeping the manuscript out of the package) or you found summaries-with-citations just work better for this kind of prescriptive content than semantic retrieval over the full prose?

Partially licensing, but also tokens + efficiency. As noted above, the book is written for humans (anecdotes, examples, etc.). Robots just need the facts and can extrapolate. If I were building a hosted service, a custom LLM with embeddings would probably beat both approaches.