CyberStrikeAI and pentagi
About CyberStrikeAI
Ed1s0nZ/CyberStrikeAI
CyberStrikeAI is an AI-native security testing platform built in Go. It integrates 100+ security tools, an intelligent orchestration engine, role-based testing with predefined security roles, a skills system with specialized testing skills, and comprehensive lifecycle management capabilities.
Based on the README, here's a technical summary that goes deeper: --- Uses native MCP (Model Context Protocol) with HTTP/stdio/SSE transports and external federation to connect AI agents directly to security tools, enabling conversational control flow through an orchestration engine that supports multi-agent delegation patterns (Eino DeepAgent). Includes vector-search knowledge base, attack-chain graph replay with risk scoring, WebShell management for post-exploitation, and optional Burp Suite integration via plugin architecture; persists all audit trails and task queues in SQLite with password-protected web UI. --- **Word count: ~65 | Key technical details**: MCP protocol variants, multi-agent orchestration, vector search, attack graphs, WebShell C2 capabilities, plugin extensibility, SQLite backend
About pentagi
vxcontrol/pentagi
✨ Fully autonomous AI Agents system capable of performing complex penetration testing tasks
Leverages a multi-agent architecture with specialized AI teams for research, exploitation, and infrastructure tasks, powered by Neo4j knowledge graphs and integration with 10+ LLM providers. Operates in isolated Docker containers with built-in tools (nmap, metasploit, sqlmap) and external search APIs, storing all findings in PostgreSQL with vector embeddings for intelligent memory and context retrieval. Provides REST/GraphQL APIs, real-time Grafana/Prometheus monitoring, and comprehensive vulnerability reporting—designed as a self-hosted, horizontally scalable microservices platform.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work