OpenKB is an open-source CLI-based knowledge base system that compiles documents into a wiki, enabling long-document handling with PageIndex-based retrieval and a bundled web UI.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
OpenKB (Open Knowledge Base) is an open-source system (in CLI) that compiles raw documents into a structured, interlinked wiki-style knowledge base using LLMs, powered by PageIndex's vectorless, reasoning-based retrieval for long documents.
How it works
When you add documents, the LLM generates a summary page, reads existing concept and entity pages, creates or updates concepts with cross-document synthesis, creates or updates entity pages, and updates the index and log. For long PDFs, PageIndex builds a tree index and the LLM reads the tree instead of the full text. The wiki is compiled and kept in sync with cross-links and summaries. Generators (query/chat/Skill Factory) produce outputs from the wiki.
Functionally, OpenKB supports: converting multiple formats (PDF, Word, Markdown, PowerPoint, HTML, Excel, CSV, text, URLs) into wiki pages; multi-modality support for figures and tables; persistent sessions for chat; exporting outputs via generators; and a Web UI bundled with a REST API.
Getting started
Install options include:
pip install openkb
Other install options:
pip install git+https://github.com/VectifyAI/OpenKB.git
git clone https://github.com/VectifyAI/OpenKB.git
cd OpenKB
pip install -e .
Quick Start example:
# 1. Create a directory for your knowledge base
mkdir my-kb && cd my-kb
# 2. Initialize the knowledge base
openkb init
# 3. Add documents
openkb add paper.pdf
openkb add ~/papers/
openkb add https://arxiv.org/pdf/2509.11420
# 4. Ask a question
openkb query "What are the main findings?"
# 5. Or chat interactively
openkb chat
Web UI setup:
pip install "openkb[web]"
openkb-web
Open http://127.0.0.1:7566/ for the Workbench.
Recent releases
Latest releases include v0.5.0-rc1 (2026-07-20) with OpenKB Studio web UI + REST API in preview, and multiple 0.4.x hotfixes and RCs focused on stability and bug fixes. Specific notes:
- v0.5.0-rc1: Web UI + REST API in preview.
- v0.4.5: Hotfix to avoid OpenAI InputTokensDetails breaking change.
- v0.4.4: Various fixes and feature notes for converter and mutation routing.
- v0.4.4-rc1: Same changes as v0.4.4.
- v0.4.3: Lint and compiler fixes.
Traction
- GitHub stars: 3252
Behind the repo
- Repository: VectifyAI/OpenKB, license Apache-2.0, language Python.
- Created: 2026-04-04; last_push: 2026-07-22.
- Topics include: agents, ai, knowledge-base, llm, rag, retrieval.
Caveats
- Dependencies include LiteLLM providers; configuration examples show model settings in .openkb/config.yaml with fields like model, language, pageindex_threshold. Examples indicate settings for provider/model, language, and PageIndex thresholds.
- README excerpts mention multiple install options and web UI setup, with details about authentication (OPENKB_API_TOKEN) and optional frontend development steps.