VectifyAI

VectifyAI / OpenKB

Public

OpenKB: Open LLM Knowledge Base

pageindex.ai agents ai knowledge-base llm rag
18
1
100% credibility
Found Apr 11, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

OpenKB is an open-source command-line tool that uses large language models to compile documents of various formats into a structured, interlinked Markdown wiki knowledge base with reasoning-based retrieval for long documents.

How It Works

1
📚 Discover OpenKB

You hear about a helpful tool that turns your scattered documents into a smart, connected wiki full of insights.

2
🛠️ Set up the tool

You easily install the program on your computer and connect your favorite AI thinking service.

3
🆕 Create your wiki space

You make a new folder where your personal knowledge base will live and grow.

4
📁 Add your documents

You drop in PDFs, articles, notes, or even whole folders, and the tool reads them to build summaries and link ideas together.

5
Explore your knowledge
❓
Ask questions

Type a question and get clear answers drawn from everything you've added.

🩺
Check health

Run a quick scan to spot loose ends or gaps in your connections.

👀
Auto-update

Turn on watch mode so new files added later update everything automatically.

🎉 Knowledge magic unlocked

You now have a living wiki that connects your ideas, answers queries, and keeps growing effortlessly.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is OpenKB?

OpenKB is a Python CLI tool that ingests documents like PDFs, Word files, PowerPoints, and HTML into an open LLM knowledge base—a persistent, wiki-style repo of summaries, concepts, and cross-links. It ditches traditional RAG's per-query rediscovery for one-time LLM compilation into Markdown files you can browse in Obsidian, query naturally, or lint for gaps. Handles long docs without vector DBs using tree-based retrieval, supports multi-modality for images/tables, and auto-updates via watch mode.

Why is it gaining traction?

Unlike basic RAG, OpenKB builds compounding knowledge: concepts synthesize across docs, contradictions get flagged, and everything links via wikilinks—no repeated LLM tax per question. Standout CLI commands like `openkb add dir/`, `openkb query "findings?"`, and `openkb lint` make it dead simple, with LiteLLM for any provider and Obsidian graph views for exploration.

Who should use this?

Researchers wrangling paper collections (think openkbp challenge datasets), engineering teams turning specs into queryable openkbs, or solo devs maintaining project knowledge bases from mixed-format docs.

Verdict

Promising alpha for Python LLM tinkerers (18 stars, 1.0% credibility score), with excellent README and Apache license, but lacks tests and scale for production—prototype your openkbs here, then harden.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.