shuck-file

Converts Word, PDF, Excel, PowerPoint, and CSV files into clean Markdown, with smart routing that depends on file size. Small documents get converted directly. Large ones return a document map showing sections, token counts, and density metrics, so your agent can pull only what it needs with targeted section extraction or keyword search. The MCP server exposes two tools: shuck for conversion with options like sections filtering, grep search, and token budgets, and list_formats for checking supported file types. Useful when you need an agent to work with office documents without dumping entire 50 page reports into context, or when you want structured extraction from spreadsheets and presentations.