Security infrastructure your AI can't be — deterministic, current past your model's training cutoff, whole-repo-aware, author-independent. Security MCP for vibe coding. 530 rules, 39 tools, CLI + doctor. Prompt-level shift-left security (secure_prompt — e

Claude Code plugin for Java JAR security audit — 基于 jar-analyzer 的 Claude Code 安全审计插件，构建数据库，AI 深入分析