topoteretes/cognee

Skip(New discovery — enrichment pending)

Why this repo matters

Automates knowledge graph construction from unstructured data to improve RAG accuracy and context retrieval.
Provides modular pipelines for entity extraction, relationship mapping, and semantic memory management in Python.
Gaining traction as developers seek structured alternatives to naive vector search for complex LLM applications.

Key Metrics

Stars: 31.3k
Forks: 3.2k
Open Issues: 458
Stars (7d delta): 908
Stars (30d delta): 908

Scores & Metadata

Trend Velocity Score: 71.92
Opportunity Score: 74.30
Confidence Score: 100.00
Language: Python
First Seen: 2 months ago

Star History

31.3k stars+1.7k over 13 snapshots8/2/2026 – 10/4/2026

Links & Sources

Related Repositories

codecrafters-io/build-your-own-x

547.5kMarkdown

freeCodeCamp/freeCodeCamp

freeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.

456.9kTypeScript

affaan-m/ECC

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

269kJavaScript

multica-ai/andrej-karpathy-skills

A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

217.5k