Read PDFs and Word documents
Link documents reads the text of PDFs and Word documents for your agent's memory. It's a plugin of its own, off until you add it, so a Link that doesn't need it never installs a document parser.
lnk plugin add link-documentsWith it, memory finds a PDF (.pdf) or a Word document (.docx) in
memory/ by the text inside it, up to 32 MiB and the first 2 MiB of
its text. Without it, memory finds one by its name alone, and a
search's reply names the plugin that reads it. Once you add it, your
agent's next start reads every document memory found by its name for
want of it.
How it runs
Memory runs lnk-link-documents text <pdf|docx> for each document, the
document on its input and the text, or why there's none, on its output.
It runs as a child of memory that holds none of memory's secrets and no
environment, with limits on its memory and its time; one that crashes
or takes over 30 seconds fails that document alone. Memory finds it
beside its own program, and its sandbox reads it where it's installed.
Its parsers
PDFs are read with lopdf, and Word documents with Link's own reader of
their text part, the zip unpacked with zip. A scanned PDF has no text
to read.