Skip to content

Complete Python3 rewrite of 'gpthidra' with support for GPT4+

Notifications You must be signed in to change notification settings

wowitsjack/Ghidra-ChatGPT

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 

Repository files navigation

Ghidra-ChatGPT

Ghidra-ChatGPT is a specialized Ghidra plugin designed to utilize the machine learning capabilities of OpenAI's GPT for semantic analysis of decompiled functions. This provides a nuanced understanding of codebase functionality.

Example Illustration

Prerequisites

Installation Procedure

  1. Acquire Script: Download the Python script Ghidra-ChatGPT.py.
  2. Invoke Script Manager: Within Ghidra, go to Window > Script Manager.
  3. Instantiate New Script: Click New, select Python as the language, and designate the script name as Ghidra-ChatGPT.py.
  4. Incorporate Code: Copy the code from Ghidra-ChatGPT.py into the editor. Substitute API_KEY = '' with the acquired OpenAI API key.
  5. Persist Script: Save the script by clicking the Save button.
  6. Activate in Tools Menu: In the Script Manager, locate your newly saved script and tick the 'In Tools Menu' checkbox.

Operational Guidelines

To invoke Ghidra-ChatGPT, adhere to one of the following procedures:

  1. Keyboard Shortcut: Execute Ctrl + Alt + G (modifiable within the script).

    - OR -

  2. Menu Navigation: Traverse to Tools -> Ghidra-ChatGPT.

Upon activation, the plugin will conduct semantic analysis on the selected function and output the findings to the Ghidra console.

Contributions

  • Original script by evyatar9
  • Port to Python 3 executed by wowitsjack

Contribution Guidelines

To contribute to the Ghidra-ChatGPT project, kindly submit pull requests or report issues through the Ghidra-ChatGPT GitHub repository.

Referential Material

About

Complete Python3 rewrite of 'gpthidra' with support for GPT4+

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%