/awesome-llm-cybersecurity-tools

A curated list of large language model tools for cybersecurity research.

Awesome Large Language Model Tools for Cybersecurity Research

Reverse Engineering

  • G-3PO: A Protocol Droid for Ghidra: An AI assistant developed by Olivia Lucca Fraser at Tenable for analysing and annotating decompiled code in Ghidra, which queries OpenAI and/or Anthropic's language models. See this writeup on the Tenable tech blog for details.
  • ai for Pwndbg: Your trusty AI debugging sidekick, developed by Olivia Lucca Fraser at Tenable as a Pwndbg command.
  • ai for GEF: Same as above, but implemented as a GEF command. Developed by Olivia Lucca Fraser at Tenable.
  • Gepetto: An IDA Pro plugin that queries GPT models for explanatory comments and meaningful variable names (like G-3PO for IDA Pro). Developed by Ivan Kwiatkowski.
  • GPT-WPRE: Whole-program Reverse Engineering with GPT-3. This is a little toy prototype of a tool that attempts to summarize a whole binary using GPT-3 (specifically the text-davinci-003 model), based on decompiled code provided by Ghidra. Developed by Brendan Dolan-Gavitt.
  • IATelligence: IATelligence is a Python script that extracts the Import Address Table (IAT) from a PE file and uses OpenAI's GPT-3 model to provide details about each Windows API imported by the file. The script also searches for related MITRE ATT&CK techniques and explains how the API could potentially be used by attackers. Developed by Thomas Roccia.

Network Analysis

  • Burp Extension for GPT: A BurpSuite plugin, developed by Tenable, that uses GPT to analyse HTTP requests and responses. Developed by Yossi Nisani at Tenable.

Cloud Security

  • EscalateGPT: Uses GPT to discover privilege escalation vulnerabilities in misconfigured Identity Access and Management (IAM) policies for AWS. Developed by Yossi Nisani at Tenable.

Proofs of Concept

Hacking LLMs

LLM-Driven Malware

  • LLMorphism: A self-replicating agent that uses GPT-3.5 as a metamorphic engine, by Second Part to Hell.
  • Darwin-GPT: A minimal self-replicating agent based on GPT-3.5/4, by Bernhard Mueller.