SkillRay — AI Skill Security Scanner. Lightweight, offline, multi-engine static analyzer for AI agent skills. 5 engines (Regex + AST + Entropy + Dataflow + Prompt), 37+ rules across 9 threat categories, Rich terminal UI, JSON/SARIF output, Claude Code native skill. No ML, no API keys — just pip install skillray.
npx add-skill MRT-8/SkillRaySkillRay — AI Skill Security Scanner. Lightweight, offline, multi-engine static analyzer for AI agent skills. 5 engines (Regex + AST + Entropy + Dataflow + Prompt), 37+ rules across 9 threat categories, Rich terminal UI, JSON/SARIF output, Claude Code native skill. No ML, no API keys — just pip install skillray.
Claude Code skills for building Apps in Toss (앱인토스) mini-apps. TDS design, review checklist, project scaffold, and policy compliance.
Production-ready AI agent skills for Claude Code, Codex, Cursor, and the Agent Skills Open Standard
Validate causal inference results with empirical permutation tests — companion to causal-impact-campaign
Build, run, and manage agent platforms.
MCP server for macOS desktop automation. Background GUI testing, 30 tools, audio capture, camera, virtual desktops. Built in Rust.
Aurelius: frontier-tier agentic LLM platform | 1.4B transformer | 146 build cycles | 25265+ tests | Python 3.14 | MCTS reasoning, BM25+hybrid search, agentic tools, multimodal, MCP, deployment | security-hardened: 0 open HIGH/CRITICAL findings