ergo-pe-av - 🧠 🦠 An artificial neural network and API to detect

项目说明

An artificial neural network and API to detect Windows malware, based on Ergo and LIEF.

Installation

cd /path/to/ergo-pe-av
sudo pip3 install -r requirements.txt

Use as an API

ergo serve /path/to/ergo-pe-a "clean, malware"

From the client, to scan a file that the server can access too:

curl "
# or
curl --data "x=/path/to/file.exe" "

To upload the whole file:

curl -F "x=@/path/to/file.exe" "

To encode a file to a of raw features:

curl -F "x=@/path/to/file.exe" "

To scan a of raw features:

curl --data "x=0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.9,......" "

Model Statistics

The dataset is made of ~200000 samples divided in two subfolders:

The dataset.csv training file has been generated with:

ergo encode ergo-pe-a/e/4TB/datapath-pe/classes --filter "*.exe"

| Training | ROC/AUC |
|----------|---------|
![]() | ![]() |

| Training | | Testing |
|----------|------------|---------|
![]() | ![]() | ![]() |

License

Made with ♥ by the dev team and it is released under the GPL 3 license.

项目信息

开发语言：Python

Star 数：182

技术要点

该项目属于病毒分析方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕🧠 🦠 An artificial neural network and API to detect Windows malware, based on Ergo and LIEF.展开，提供了可运行的实现，对理解相关安全机制、工具原理与攻防技术具有较高的研究价值。

从技术实现角度看，此类项目的核心在于对底层原理的深入理解与工程化落地。开发者在研究过程中可以结合官方文档与社区讨论，梳理项目的初始化流程、核心模块职责与数据流转，逐步掌握其中的技术细节与设计思路。

使用与学习建议

1. 下载源码压缩包，解压到本地环境。

2. 根据项目类型准备对应的运行环境与依赖。

3. 按照 README 中的步骤完成配置与部署。

4. 在隔离的测试环境中运行验证，避免影响生产系统。

免责声明

本资源来源于开源社区，仅供安全研究、技术学习与合法授权范围内的攻防演练使用，请勿用于任何非法或侵害他人权益的用途。使用本项目产生的一切后果由使用者自行承担。