SujaaViswanathan/Modern_Cyber_Threat_Simulation_Dataset
收藏Hugging Face2026-03-22 更新2026-03-29 收录
下载链接:
https://hf-mirror.com/datasets/SujaaViswanathan/Modern_Cyber_Threat_Simulation_Dataset
下载链接
链接失效反馈官方服务:
资源简介:
---
license: mit
task_categories:
- text-generation
language:
- en
tags:
- cybersedcurity
- scripts
- blockchain
- IOT
pretty_name: sunnythakur
size_categories:
- n<1K
---
Modern Cyber Threat Simulation Dataset
Overview
The Modern Cyber Threat Simulation Dataset is a comprehensive collection of 200 simulated cyber threats, vulnerabilities, and exploits tailored for 2025's advanced technological landscape. Covering AI/ML, Blockchain, Cloud, and IoT domains, this dataset provides vulnerable code/configurations, fuzzing-based exploit scripts, mitigations, and AI training prompts to support cybersecurity research, red teaming, and defensive tool development. Each entry is designed to simulate real-world attack vectors, emphasizing unconventional tactics and fuzzing to uncover edge cases.
Purpose
This dataset aims to:
Enable red teamers to simulate advanced cyber threats and test system resilience.
Support data scientists in training AI/ML models for threat detection and mitigation.
Provide security researchers with a structured resource for analyzing vulnerabilities and developing countermeasures.
Foster innovation in pentesting tools by offering exploit scripts in diverse languages (Python, Go, Rust, JavaScript, Bash, etc.).
```
Dataset Structure
The dataset is organized as JSON artifacts, with entries 1–200 split across multiple files:
Entries 1–137: Initial threats (AI/ML, Blockchain).
Entries 138–170: Extended AI/ML and Blockchain threats.
Entries 171–200: Expanded to include Cloud and IoT threats.
```
```
Each entry contains:
ID: Unique identifier.
Category: AI/ML, Blockchain, Cloud, or IoT.
Threat: Specific attack vector (e.g., Model Weight Tampering, Smart Contract Reentrancy).
Description: Brief overview of the threat.
Code/Config: Vulnerable code or configuration snippet.
Exploit Script: Fuzzing-based script to demonstrate the exploit.
Mitigation: Code or configuration to address the vulnerability.
References: Links to OWASP, CWE, and MITRE ATT&CK frameworks.
AI Training Prompt: Guidance for training AI models to detect or mitigate the threat.
```
```
Example Entry
{
"id": 138,
"category": "AI/ML",
"threat": "Model Weight Tampering",
"description": "Tampering with ML model weights via fuzzing serialized models, altering predictions.",
"code_config": "import joblib\nmodel = joblib.load('model.pkl')\ndef predict(data):\n return model.predict(data)",
"exploit_script": "...",
"mitigation": "...",
"references": ["OWASP Top 10 for LLM...", "CWE-502...", "MITRE ATT&CK T1055..."],
"ai_training_prompt": "Fuzz ML model weights to detect tampering and suggest cryptographic signing."
}
```
Usage
Prerequisites
Languages: Python, Go, Rust, JavaScript, Bash, Solidity, YAML.
Tools: TensorFlow, Flask, Web3.py, Boto3, Kubernetes client, MQTT, CoAP, Z-Wave, BLE libraries.
Environment: Local blockchain nodes (e.g., Ganache), AWS/GCP cloud accounts, IoT simulation setups.
Explore Entries:
Navigate to data/ for JSON artifacts.
Use scripts/ for exploit and mitigation scripts.
Run Exploits:
Follow entry-specific instructions to set up vulnerable environments.
Execute exploit scripts with caution in isolated environments.
Apply Mitigations:
Implement provided mitigations to secure systems.
Train AI Models:
Use AI training prompts to develop detection models with frameworks like TensorFlow or PyTorch.
Example Use Case
To simulate a Smart Contract Reentrancy attack (ID 151):
Deploy the vulnerable contract on a local Ethereum node.
Run the provided JavaScript exploit script to drain funds.
Apply the mitigation (reentrancy guard) and retest.
Contribution Guidelines
We welcome contributions to expand the dataset or improve existing entries. To contribute:
Fork the repository.
Create a new branch (feature/new-threat or fix/entry-123).
Add new entries or update existing ones, ensuring:
Unique threats with no overlap.
Fuzzing-based exploits in diverse languages.
References to OWASP, CWE, and MITRE ATT&CK.
Submit a pull request with a clear description of changes.
License
This dataset is licensed under the MIT License. Use it freely for research and development, but ensure compliance with ethical and legal standards.
Contact
For questions or collaboration, contact the sunny48445@gmail.com.
Acknowledgments
Inspired by OWASP Top 10, CWE, and MITRE ATT&CK frameworks.
Built by sunny thakur .
---
许可证:MIT许可证
任务类别:
- 文本生成
语言:
- 英语
标签:
- 网络安全
- 脚本
- 区块链
- 物联网(IoT)
数据集展示名称:sunnythakur
规模类别:
- 样本数少于1000(n<1K)
---
现代网络威胁模拟数据集
概述
现代网络威胁模拟数据集是一套覆盖2025年先进技术场景的综合性数据集,包含200组模拟网络威胁、脆弱点与利用手段。该数据集涵盖人工智能/机器学习(AI/ML)、区块链、云计算与物联网(IoT)领域,提供存在漏洞的代码/配置文件、基于模糊测试的利用脚本、缓解方案,以及用于大语言模型(LLM)训练的提示词,可支撑网络安全研究、红队演练与防御工具开发。每一条数据均旨在模拟真实世界的攻击路径,重点关注非常规战术与模糊测试方法,以挖掘边缘场景漏洞。
目标
本数据集旨在达成以下目标:
1. 协助红队人员模拟高级网络威胁,测试系统韧性;
2. 支持数据科学家训练用于威胁检测与缓解的AI/ML模型;
3. 为安全研究人员提供结构化资源,用于分析脆弱点并开发应对措施;
4. 提供多语言(Python、Go、Rust、JavaScript、Bash等)的利用脚本,推动渗透测试工具的创新发展。
数据集结构
本数据集以JSON格式文件组织,1-200条数据分布于多个文件中:
- 条目1-137:初始威胁(覆盖人工智能/机器学习(AI/ML)、区块链领域);
- 条目138-170:扩展AI/ML与区块链威胁;
- 条目171-200:新增云计算与物联网(IoT)威胁。
单条数据字段说明
每条数据包含以下字段:
- ID:唯一标识符;
- 类别:AI/ML、区块链、云计算或物联网;
- 威胁类型:具体攻击路径(例如:模型权重篡改、智能合约重入攻击);
- 描述:威胁的简要概述;
- 代码/配置:存在漏洞的代码或配置片段;
- 利用脚本:用于演示漏洞利用的模糊测试脚本;
- 缓解方案:用于修复该脆弱点的代码或配置;
- 参考资料:指向OWASP、CWE与MITRE ATT&CK框架的链接;
- AI训练提示词:用于训练AI模型检测或缓解该威胁的指导提示。
示例条目
{
"ID": 138,
"类别": "AI/ML",
"威胁类型": "模型权重篡改",
"描述": "通过模糊测试序列化模型的方式篡改机器学习模型权重,从而改变模型预测结果。",
"代码/配置": "import joblib
model = joblib.load('model.pkl')
def predict(data):
return model.predict(data)",
"利用脚本": "...",
"缓解方案": "...",
"参考资料": ["OWASP Top 10 for LLM...", "CWE-502...", "MITRE ATT&CK T1055..."],
"AI训练提示词": "针对模型权重篡改进行模糊测试,以检测篡改行为并提出加密签名建议。"
}
使用指南
前置条件
- 编程语言:Python、Go、Rust、JavaScript、Bash、Solidity、YAML;
- 工具库:TensorFlow、Flask、Web3.py、Boto3、Kubernetes客户端、MQTT、CoAP、Z-Wave、蓝牙(BLE)相关库;
- 运行环境:本地区块链节点(例如Ganache)、AWS/GCP云账号、物联网模拟环境。
浏览数据条目
- 前往data/目录查看JSON格式数据文件;
- 使用scripts/目录下的脚本运行利用与缓解代码。
执行漏洞利用
- 按照单条数据的专属说明搭建存在漏洞的运行环境;
- 请在隔离环境中谨慎执行利用脚本。
应用缓解方案
- 按照提供的缓解方案加固系统。
训练AI模型
- 使用提供的AI训练提示词,结合TensorFlow或PyTorch等框架开发威胁检测模型。
示例用例
以模拟智能合约重入攻击(条目ID 151)为例:
1. 在本地以太坊节点部署存在漏洞的合约;
2. 运行提供的JavaScript利用脚本转移资金;
3. 应用重入保护缓解方案后重新进行测试。
贡献指南
我们欢迎各类贡献,以扩展数据集内容或优化现有条目。贡献流程如下:
1. Fork本代码仓库;
2. 创建新分支(例如feature/new-threat或fix/entry-123);
3. 添加新条目或更新现有条目,需确保:
- 威胁类型唯一且无重复;
- 采用多语言编写的基于模糊测试的利用脚本;
- 引用OWASP、CWE与MITRE ATT&CK相关资料;
4. 提交清晰描述变更内容的拉取请求。
许可证
本数据集采用MIT许可证进行授权。可自由用于研究与开发,但需遵守伦理与法律标准。
联系方式
如有疑问或合作意向,请联系 sunny48445@gmail.com。
致谢
本数据集灵感来源于OWASP Top 10、CWE与MITRE ATT&CK框架;由sunny thakur 制作。
提供机构:
SujaaViswanathan


