AI-powered bug bounty hunting from your terminal, this project provides a co-pilot that reasons about security testing, prioritizes targets by payout likelihood, and orchestrates over 25 tools. It addresses the need for a more intelligent, integrated approach to bug bounty hunting, moving beyond collections of disparate scripts to offer an end-to-end automated pipeline from reconnaissance to report generation. The target audience is security researchers and bug bounty hunters looking to enhance their efficiency and effectiveness.

How It Works

The project integrates Claude Code with a multi-stage pipeline: Reconnaissance, Learning (intelligence gathering), Hunting (scanning), and Reporting. Claude orchestrates this flow, understanding security methodologies rather than just executing commands. The core advantage lies in its AI co-pilot capability, which intelligently maps attack surfaces, prioritizes testing based on potential payout, and drives a suite of security tools in the correct sequence. Users can run the entire pipeline or execute individual stages independently.

Quick Start & Requirements

• Primary install: Clone the repository, make install_tools.sh executable, and run it. Copy SKILL.md to ~/.claude/skills/bug-bounty/SKILL.md.
• Prerequisites: macOS: brew install go python3 node jq. Linux (Debian/Ubuntu): sudo apt install golang python3 nodejs jq. Requires Python 3.8+ and API keys for Claude. The installation script installs 18+ security tools including subfinder, nuclei, sqlmap, and dalfox.
• Links: GitHub Repo, Claude Code Skill.

Highlighted Details

• AI co-pilot that reasons about testing, prioritizes by payout likelihood, and drives 25+ tools.
• Comprehensive methodology embedded in a 1,200+ line SKILL.md file, covering recon, vulnerability classes, bypass tables, and AI/LLM hunting frameworks.
• Supports a wide array of vulnerability classes including IDOR, SSRF, XSS, SQLi, OAuth, Race Conditions, LLM injection, and Web3 security.
• Generates submission-ready reports for platforms like HackerOne, Bugcrowd, and Intigriti.

Maintenance & Community

The project is maintained by shuvonsec. Contact is available via GitHub, Email, Twitter, and LinkedIn.

Licensing & Compatibility

The project is released under the MIT License, which is permissive for commercial use and integration into closed-source projects.

Limitations & Caveats

The tool is intended for authorized security testing only, requiring explicit permission and adherence to program rules of engagement. It relies on the capabilities of the Claude AI model and the correct configuration of API keys.