PentAGI
Open-source autonomous penetration testing agents that run 20+ security tools in an isolated Docker sandbox with memory and web intelligence.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for PentAGI, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
PentAGI (vxcontrol/pentagi) is an open-source autonomous AI agent system for penetration testing. It plans and executes complex offensive security workflows end-to-end, running professional pentesting tools (e.g., nmap, Metasploit, sqlmap and others) inside a sandboxed Docker environment for isolation. PentAGI includes multi-step task orchestration, a browser-based intelligence component for gathering current info, and a smart memory system to store findings and successful approaches for future assessments. It is intended for authorized security testing environments where you have explicit permission to run assessments.
Key features and capabilities
- penetration testing
- red teaming
- docker sandbox
- nmap
- metasploit
- sql injection testing
- web intelligence
- tool orchestration
- long-term memory
- authorized security testing
Use cases
- Automating reconnaissance, enumeration, and exploitation planning steps during authorized penetration tests.
- Running pentesting tools safely in an isolated Docker sandbox to reduce host risk.
- Generating repeatable testing workflows with stored findings and reusable approaches via long-term memory.
- Augmenting security assessments with built-in web intelligence for up-to-date techniques and references.