redamon

Active
GitHub Python MIT

Description

An AI-powered agentic red team framework that automates offensive security operations, from reconnaissance to exploitation to post-exploitation, with zero human intervention.

Key Features

  • Autonomous offensive pipeline chaining recon, exploitation, and post-exploitation into a single workflow with human oversight at each critical step
  • Fireteam parallel multi-agent architecture — multiple AI agents test different attack vectors concurrently using Hydra, CVE exploits, and XSS scanning
  • Neo4j knowledge graph that merges all tool outputs into a deduplicated, queryable attack surface with explicit relationships
  • CypherFix AI triage agent correlates findings, ranks by exploitability, then CodeFix agent clones repos, implements fixes, and opens GitHub PRs
  • 100+ integrated security tools including Metasploit, Nmap, Nuclei, SQLMap, Hydra, and OpenVAS with 185,000+ detection rules
  • Configurable autonomy with RoE guardrails — supports local models via Ollama, vLLM, and LM Studio with 400+ AI model options

Use Cases

💡 Automated penetration testing for authorized security teams running full-scope assessments against target infrastructure
💡 CI/CD security gate that continuously scans repositories and auto-remediates vulnerabilities via pull requests
💡 Security research and CVE validation — autonomously verify exploitability of newly disclosed vulnerabilities
💡 Educational red team lab for learning offensive security techniques with full attack chain visibility
💡 Compliance auditing — map entire attack surfaces and generate AI-driven pentest reports with remediation code

Strengths & Limitations

Strengths

  • Actively maintained, recent updates
  • High community interest (2.3k stars)
  • Permissive open-source license (MIT)
  • Responsive to issues, low backlog

Quick Start

Clone the repo, install via Docker Compose, configure your target scope and AI provider API key in the project settings (500+ configurable options), then run the pipeline. RedAmon performs 6-phase recon, hands off to exploitation agents, triages findings via CypherFix, and opens a GitHub PR with code fixes — all with configurable autonomy levels.

Related Projects