Add synthetic HTTP attack dataset with structured examples and tooling

[Copilot]

Creates a structured dataset of HTTP request/response pairs demonstrating common web security attacks for research, ML training, and security tool validation.

Dataset Structure

• 6 attack categories: SQL injection, XSS, CSRF, path traversal, command injection, XXE
• 8 examples with complete HTTP details, attack vectors, payloads, and detection indicators
• JSON schema for validation and programmatic access
• Index file with metadata and statistics

Usage

Load and analyze the dataset:

import json
from pathlib import Path

def load_dataset(dataset_path='dataset'):
    examples = []
    for category_dir in Path(dataset_path).iterdir():
        if category_dir.is_dir():
            for example_file in category_dir.glob('*.json'):
                try:
                    with open(example_file, 'r') as f:
                        examples.append(json.load(f))
                except json.JSONDecodeError as e:
                    print(f"Error parsing {example_file}: {e}")
    return examples

dataset = load_dataset()
# Access attack details: method, URL, headers, body, payload, indicators

Example Entry

Each attack includes:

• HTTP request/response (method, URL, headers, body, status)
• Attack vector explanation
• Malicious payload extraction
• Detection indicators for security tools

JavaScript examples also provided in examples/load_dataset.js.

Original prompt

this is repository to store a dataset of examples of http requests and responses of malicious attacks in several categories

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.