# RKnowledge

> Production-grade knowledge graph extraction CLI. Extract concepts and relationships from any document using LLMs, store in Neo4j, and explore with interactive visualization.

## About

RKnowledge is a Rust CLI that transforms documents into a queryable knowledge graph. It supports multiple LLM providers, parallel extraction, incremental updates, and graph analytics. Inspired by rahulnyk/knowledge_graph (Python), rewritten from scratch with 118 tests.

## Install

Quick install: `curl -fsSL https://raw.githubusercontent.com/Algiras/RKnowledge/main/install.sh | sh`
From source: `cargo install --path .`
Prebuilt binaries: https://github.com/Algiras/RKnowledge/releases

## Core Commands

- `rknowledge extract <file>` - Extract concepts and relationships from a document
- `rknowledge query <term>` - Query the knowledge graph
- `rknowledge path <from> <to>` - Find shortest path between concepts
- `rknowledge stats` - Graph statistics (PageRank, communities, density)
- `rknowledge communities` - Detect concept clusters
- `rknowledge visualize` - Launch interactive graph visualization

## Features

- **Multi-format**: PDF, Markdown, HTML, plain text
- **Multi-provider LLM**: Anthropic, OpenAI, Google, Ollama
- **Parallel extraction**: Concurrent LLM calls with `-j` flag
- **Neo4j backend**: Persistent graph DB, incremental `--append`
- **Graph analytics**: PageRank, community detection, shortest path
- **Export**: JSON, CSV, GraphML, Cypher
- **Tenant isolation**: Multiple projects in one Neo4j instance

## Source

- Repository: https://github.com/Algiras/RKnowledge
- Documentation: https://algiras.github.io/RKnowledge/
- License: MIT
- Language: Rust