root-kg
Active·★ 16·MIT·Updated 2026-10-12
ROOT is a local-first knowledge management system that connects your personal notes, meetings, and emails into a queryable intelligence layer. It builds an LLM-powered knowledge graph to provide cited, natural language answers to complex questions, outperforming traditional keyword search.
#Knowledge Graph#RAG#Semantic Search#Obsidian Integration#LLM Orchestration#Entity Extraction#SQLite Database
01
Features
01LLM-powered entity and relation extraction to build a comprehensive knowledge graph.
02Multi-source data ingestion from Obsidian vaults, meeting notes, and emails.
03GraphRAG for synthesized, cited natural language answers, combining semantic search with graph traversal.
04Local-first architecture with low running costs, utilizing a single SQLite file for all data.
05Incremental indexing and extraction, only reprocessing changed notes for efficiency and cost savings.
02
Compatibility
Python
Runtime
Verified via docs
Anthropic
LLM
Verified via docs
OpenRouter
LLM
Verified via docs
Ollama
LLM
Verified via docs
SQLite
Database
Verified via docs
MacOS
OS
Verified via docs
03
Quick start
1
$ git clone https://github.com/mshadmanrahman/root-kg.git
2
$ cd root-kg
3
$ python -m venv .venv
4
$ source .venv/bin/activate
5
$ pip install -e .
04
Use cases
↳For Product Managers: Trace decision evolution, identify project stakeholders, and get briefs before 1:1 meetings.
↳For Engineers: Understand system architecture, identify service dependencies, and find stale topics.
↳For Researchers & Writers: Discover patterns via clustering, identify knowledge gaps, and connect cross-domain links.
↳For Teams: Generate weekly activity summaries and perform project health checks across all sources.
05
Alternatives
mcp-local-rag★ 416
Local-first RAG server for developers. Semantic + keyword search for code and technical docs. Works with MCP or CLI. Fully private, zero setup.
knowledge-graph-system★ 129
Kappa Graph — κ(G). A semantic knowledge graph where knowledge has weight. Extracts concepts, measures grounding strength, preserves disagreement, traces everything to source.
crg★ 76
Knowledge graph for token-efficient code reviews -- semantic search and call-graph resolution across your codebase.
kglite★ 73
Embedded Cypher knowledge graph for Python and Rust. Bundled MCP server, describe() schema, and code-graph parser for LLM agents.
Comments
Log in to leave a comment
No comments yet. Be the first!