Info Disclosure Scanner.

AI-powered scanner for information disclosure vulnerabilities

GitHub
Info Disclosure Scanner architecture

Overview

An AI-powered security scanner that detects sensitive information leakage in web applications using Claude AI. Finds exposed .git repos, .env files, database dumps, admin panels, stack traces, and secrets in JavaScript.

What It Detects

SeverityExamples
HighExposed .git repos, .env files, database dumps, hardcoded passwords
MediumAdmin panels, stack traces, secrets in JavaScript, debug endpoints
LowServer version disclosure, robots.txt, technology fingerprinting
InfoMissing security headers (CSP, HSTS, X-Frame-Options)

Demo with OWASP Juice Shop

docker run -d -p 3000:3000 bkimminich/juice-shop
python scanner.py http://localhost:3000 --html report.html
# Finds 80+ vulnerabilities

How It Works

1

Scan

Checks 50+ sensitive paths, response headers, JavaScript files, and error responses for information leakage.

2

AI Analysis

Claude AI analyzes findings to identify real information disclosure vulnerabilities with severity ratings.

3

Report

Generate HTML reports with risk scores and detailed findings. Also supports Markdown, JSON, and SARIF output.

Tech Stack

PythonClaude APIHTML Reports