Decepticon is an AI-powered autonomous red team testing framework designed to simulate realistic attack chains for cybersecurity defense.
Source: README View on GitHub →Decepticon is gaining attention due to its unique approach to red teaming, leveraging AI to execute complex attack chains with discipline and isolation, addressing the gap in realistic red teaming tools and showcasing innovative use of AI in cybersecurity.
Source: Synthesis of README and project traitsDecepticon simulates realistic attack chains, including reconnaissance, exploitation, privilege escalation, lateral movement, and command and control, mimicking the behavior of real adversaries.
Source: READMEBefore any action, Decepticon generates a comprehensive engagement package with rules and objectives, ensuring disciplined operations.
Source: READMEAll commands run in a Kali Linux sandbox on a dedicated operational network, isolating the management plane from the sandbox environment.
Source: READMESupports a tier-based, credentials-aware fallback chain with various AI models and providers, allowing for flexible configuration.
Source: READMEDecepticon features a two-network design with management services on 'decepticon-net' and sandbox operations on 'sandbox-net'. It uses a knowledge graph for persistent findings and employs a Docker-based architecture for containerization and isolation.
Source: README + Code treeCenter: project; inner ring: core feature modules; outer ring: key dependencies. Auto-generated from core_features and tech_stack.key_deps.
pydanticlangchain-corelangchain-openailanggraphneo4jDecepticon is suitable for cybersecurity professionals, red teamers, and organizations looking to simulate realistic attack scenarios for defense improvement. It is useful for penetration testing, security audits, and offensive security training.
Source: READMEv1.0.24 (2026-05-09): Refactor middleware tools and harden OPPLAN persistence
Source: GitHub ReleasesDecepticon is a promising project for cybersecurity professionals seeking advanced red teaming capabilities with AI integration. It is particularly suited for teams looking to enhance their offensive security capabilities and simulate complex attack scenarios with discipline and isolation.
Source: Synthesis