pentest-ai-agents – 28 Claude Code Subagents for Penetration Testing

The document announces pentest-ai-agents, an open-source collection of 28 Claude Code subagents that specialize across the penetration-testing lifecycle (recon, web testing, AD attacks, cloud, mobile, social engineering, exploit chaining, detection engineering, forensics, malware analysis, and reporting). It details a simple installer, a two-tier agent model (advisory Tier 1 and executable Tier 2 with command approval), a SQLite-backed findings database, options for local model deployment, and an extended MCP server with tool wrappers and CI/CD integrations.