# VectifyAI/OpenKB

OpenKB: Open LLM Knowledge Base

Repository: https://github.com/VectifyAI/OpenKB
Canonical: https://ross.abutalabs.com/products/openkb
Homepage: https://pageindex.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: knowledge-base, rag, ai, llm, agents, retrieval
Last push: 2026-07-22T03:45:16+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 93, release rhythm 93, longevity 10
- inputs: {"age_days": 151, "days_push": 42, "days_rel": 44, "gap_med": 3.5, "n_releases_24m": 13}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3847, forks 413 (observed 2026-08-28T04:08:24.179163+00:00)

## What it is
OpenKB is an open-source CLI tool that compiles raw documents (PDF, Word, Markdown, HTML, and more) into a structured, interlinked wiki-style knowledge base using LLMs. It is powered by PageIndex's vectorless, reasoning-based retrieval, which uses tree indexing and agentic LLM reasoning instead of embeddings, chunking, or vector databases.

## Use cases
- build a persistent knowledge base from my pdf documents
- query long documents without a vector database
- turn a folder of documents into a wiki with cross-references
- chat with my document collection using reasoning-based retrieval
- extract entity pages for people and organizations from documents
- create agent skills from my accumulated knowledge base
- retrieve tables and figures from complex documents
- replace traditional rag with vectorless document retrieval

## When to choose
- you want knowledge to compound over time instead of being re-derived on every query
- you work with long, complex documents where chunk-based vector similarity loses context
- you need traceable, explainable retrieval grounded in source documents
- you want native multi-modal understanding of figures, tables, and images
- you prefer a persistent compiled wiki over ephemeral per-query retrieval

## When to avoid
- you need low-latency similarity search over massive embedding corpora
- you want a hosted web application rather than a CLI workflow
- your pipeline already depends on a specific vector database stack
- you need real-time indexing of rapidly changing data streams
- you require a mature stable release for production-critical systems

## Facets
- artifact type: cli-tool
- maturity: active
- function: rag, nlp, llm-inference, agent-framework, search-engine, documentation, pdf, ocr, chatbot, developer-tools
- domain: large-language-models, artificial-intelligence, developer-tools
- platform: python, cli, cross-platform
- tags: knowledge-base, vectorless-retrieval, wiki-generation, document-understanding, llm-powered, open-knowledge-format, entity-extraction, multi-modal, long-documents, skill-factory, retrieval-augmented-generation, natural-language-processing, knowledge-management, document-processing, ai-agents

## Member repositories
- VectifyAI/OpenKB (main) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:24.179163+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:25:59.827327+00:00, confidence not recorded.
  - readme: https://github.com/VectifyAI/OpenKB (fetched 2026-08-28T04:08:24.179163+00:00, sha f8559095a3b2)
  - homepage: https://pageindex.ai (fetched 2026-08-29T09:20:39.844396+00:00, sha 8c4e16931f05)
  - site_page: https://pageindex.ai/about (fetched 2026-08-29T09:20:39.853751+00:00, sha 2fee92fee4d0)
  - site_page: https://docs.pageindex.ai (fetched 2026-08-29T09:20:39.855841+00:00, sha 3ef3ed6e22e1)
  - registry_pypi: https://pypi.org/pypi/openkb/json (fetched 2026-08-29T09:20:39.857678+00:00, sha b059c732958c)
- Data as of 2026-08-30T08:39:29.467469+00:00.