kb-retriever

Searches through a folder of documents, PDFs, and spreadsheets to find the answer to a question.

Installation
Run `npx skills add "https://github.com/conardli/garden-skills" --skill "kb-retriever"` to install this skill, then follow its SKILL.md instructions for my next request.

Paste this into Claude Code, Cursor, or any agent that can run commands.

What this skill does
What it does: This skill searches through a local folder of documents, such as text files, PDFs, and spreadsheets, to answer a question. Rather than opening every file, it first reads small index files that describe what is inside each folder, then narrows down to the few files most likely to have the answer. For PDFs and spreadsheets, it always learns the proper way to extract data from that file type first, instead of guessing, so it does not waste effort or miss information. It searches step by step, looking at small relevant sections of a file rather than loading the whole thing at once, and it tries up to five rounds of searching with different keywords before giving up. If it still cannot find a good answer, it says so clearly and tells the person what to try, such as narrowing the topic or naming a more specific file, rather than making something up. When to use it: - Answering questions using a local folder of documents, reports, or spreadsheets - Looking something up in a large set of files without reading everything from start to finish

The full skill text could not be loaded right now. It installs fine with the command above, or read it at the source.

Mirrored from the author's public source. Install counts from the open skills registry.

The systems behind these skills get built for partners every week.

Partner with us