kb-retriever

Retrieve answers from local Markdown, PDF, and Excel knowledge directories.

Updated Jul 3, 2026
One-click install
npx skills add https://github.com/John198912/Content_Workflow_Garden_codex --skill kb-retriever-john198912
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kb-retriever
Source: https://github.com/John198912/Content_Workflow_Garden_codex/tree/main/claude_space/.claude/skills/kb-retriever
Command: npx skills add https://github.com/John198912/Content_Workflow_Garden_codex --skill kb-retriever-john198912

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires grep, read_file, pdftotext, pandas, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

The Kb Retriever Skill simplifies accessing information from a local, multi-format knowledge directory. It allows for efficient, question-driven exploration of large document collections without loading entire files into context.

Core Features & Use Cases

  • Multi-format Support: Handles Markdown, PDF, Excel, and more.
  • Hierarchical Indexing: Navigates structured indexes within knowledge directories.
  • Progressive Retrieval: Searches with 'grep' and limited reads, keeping token usage low.
  • Learn Before Process: Ensures PDF and Excel data are handled with precision before searching.
  • Use Case: Suppose you have a vast repository of documents including PDF reports and Excel spreadsheets. You can ask questions like "What were the sales numbers in the second quarter of last year?" and the Skill retrieves the information efficiently.

Quick Start

To retrieve information from your knowledge base, simply use the Kb Retriever Skill and ask a question like "Tell me the latest sales numbers."

Frequently Asked Questions about kb-retriever

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search for specific information across multiple PDF and Excel documents?

To search for specific information across multiple PDF and Excel documents, use progressive retrieval to navigate hierarchical indexes and selectively read files. This method uses 'grep' and limited reads to keep token usage low while efficiently extracting targeted data.

What is the best way to answer questions from a local knowledge base without loading entire files into context?

The best way to answer questions from a local knowledge base without loading entire files is progressive retrieval. It navigates hierarchical indexes and performs selective reads, ensuring low token consumption while efficiently finding the exact information needed.

Does progressive retrieval work with both Markdown and PDF formats?

Yes, progressive retrieval works with both Markdown and PDF formats, alongside Excel. It navigates hierarchical indexes within the knowledge directory and utilizes tools like 'pdftotext' to ensure precise data handling before searching.

Do I need pandas and pdftotext to perform document analysis on a local knowledge directory?

Yes, you need 'pandas' and 'pdftotext' to perform document analysis on a local knowledge directory. These dependencies are required to parse and process Excel spreadsheets and PDF files, respectively, ensuring precise data extraction.

How do I efficiently query a large repository of local PDF reports and Excel spreadsheets?

To efficiently query a large repository of local PDF reports and Excel spreadsheets, use a question-driven exploration approach. It searches with 'grep' and performs limited reads, navigating hierarchical indexes to retrieve information without loading entire files.