peng-gao-lab/ctinexus

CTINexus is a framework that leverages optimized in-context learning of LLMs to enable data-efficient extraction of cyber threat intelligence and the construction of high-quality cybersecurity knowledge graphs.

61
/ 100
Established

The framework uses a multi-stage pipeline combining optimized in-context learning with demonstration retrieval for entity extraction, hierarchical entity alignment (typing and merging with IOC protection), and link prediction to complete knowledge graphs. It supports multiple LLM providers (OpenAI, Google Gemini, AWS Bedrock, and local Ollama) with flexible configuration, exposing functionality via Python package, Gradio web interface, and CLI for processing both raw text and CTI report URLs.

Available on PyPI.

Maintenance 10 / 25
Adoption 9 / 25
Maturity 24 / 25
Community 18 / 25

How are scores calculated?

Stars

71

Forks

16

Language

Python

License

MIT

Last pushed

Feb 25, 2026

Commits (30d)

0

Dependencies

14

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/peng-gao-lab/ctinexus"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.