Get answers and structured data from document files.
document extraction
Extracts text, tables, and specific values from document files using a local parser.
When to use it
Use when a task requires reading or extracting information from a PDF, Office document, or image.
Give it a document and what to find; it extracts the relevant text, tables, or values and answers from them.
What you provide
This skill
@llamaindex/liteparse
Installs the LiteParse package
/tmp/doc.txt
Saves extracted text in /tmp
/tmp/shots/
Saves page screenshots in /tmp
Requires Node 18 or newer.
Requires the @llamaindex/liteparse package, which provides the lit CLI.
LibreOffice is required when processing Office files.
ImageMagick is required when processing images.