# CyberStrike

An open-source AI agent for automated, autonomous penetration testing from the terminal, using a developer's own Claude, GPT or other LLM subscription across 13+ specialized agents and OWASP-based test cases.

CyberStrike is an open-source offensive-security harness that turns any connected LLM into an autonomous red-team agent. It runs as a terminal TUI, asks for an LLM provider and API key on first run, and then handles reconnaissance, vulnerability discovery, exploitation and reporting from a plain description of what to test, using 13 or more specialized agents, over 7,600 security skills and more than 120 OWASP test cases.

Rather than wrapping an LLM directly, its intelligence layer injects domain-specific context, such as OWASP methodology, vulnerability patterns, attack-chain reasoning and tool-orchestration logic, into every interaction, and normalizes structured output across more than 150 AI providers and 5,300 models. It also ships a Web UI and integrates with an MCP tool ecosystem of 176 or more tools.

It can reuse an existing Claude Code or OpenAI subscription instead of requiring separate API costs.

Website: https://cyberstrike.io