Show HN: LLMtary – Local LLM Red-Teaming Tool

Category: security

Tags: security, llm, penetration-testing, red-teaming, local-ai

Score: 7.3/10 (Innovation: 7, Technical: 8, Documentation: 8, Utility: 6)

LLMtary is a local-first, LLM-driven penetration testing platform that autonomously executes multi-phase security assessments from reconnaissance through exploitation validation. It's interesting because it combines AI agentic workflows with real command execution, OPSEC-aware prompting, and structured reporting while running entirely on-premise with local models.

Target audience: security professionals, penetration testers, devops, red teams

Repository: https://www.llmtary.com/ · Dart · MIT

View on Hacker News