knowledge-base-builder

Create an AI-ready knowledge base from scattered documents with taxonomy and chunking.

2|Updated Apr 25, 2026
One-click install
npx skills add https://github.com/viethahong/business-skills --skill knowledge-base-builder-viethahong
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: knowledge-base-builder
Source: https://github.com/viethahong/business-skills/tree/main/skills/ops/knowledge-base-builder
Command: npx skills add https://github.com/viethahong/business-skills --skill knowledge-base-builder-viethahong

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

SMEs often lose critical know-how when staff leave or joiners struggle to locate scattered documents. This skill provides a structured, AI-ready internal knowledge base capable of supporting RAG-based querying and scalable information access across teams.

Core Features & Use Cases

  • Design a taxonomy and hierarchical structure to organize company documents (Notion, Obsidian, Dify) for consistent AI access.
  • Enable preprocessing, chunking, and metadata enrichment to create searchable, context-preserving content.
  • Use Case: Build a centralized knowledge base from policies, SOPs, and product docs to power AI assistants and internal search.

Quick Start

Create a structured AI-ready knowledge base from scattered documents to enable rapid, RAG-enabled AI access.

Frequently Asked Questions about knowledge-base-builder

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

FAQPage Schema
How do I build an AI-ready knowledge base from scattered company documents?

A knowledge base supports RAG workflows by applying taxonomy, hierarchical structuring, and metadata enrichment to your documents. This preprocessing ensures that chunked content retains its context, enabling accurate and secure retrieval for AI assistants.

Can I use this to organize internal policies in Notion and Obsidian for AI access?

Yes, you can organize internal policies, SOPs, and product docs across platforms like Notion and Obsidian. The skill designs a hierarchical structure and applies access controls to ensure consistent and secure AI access across your teams.

What is the best way to chunk documents for RAG workflows without losing context?

The best way to chunk documents for RAG is to combine preprocessing with metadata enrichment. By applying a structured taxonomy before chunking, the content maintains its hierarchical context, ensuring accurate retrieval when queried by AI.

Do I need to set up access controls when creating an internal knowledge base for teams?

Yes, implementing access controls is necessary when creating an internal knowledge base for teams. The skill applies security measures during the metadata enrichment process to ensure that document retrieval remains accurate and restricted to authorized personnel.

How does metadata enrichment improve internal search and Dify integration?

Metadata enrichment improves internal search and Dify integration by attaching descriptive tags and hierarchical context to chunked documents. This allows AI assistants to accurately locate and retrieve specific information from the centralized knowledge base.