Nate-Carroll-Cyber/AI-Threat-Modeling-Claude-Skill
Layered, evidence-gated security assessments for agentic AI (LLM/RAG/MCP/multi-agent) using MAESTRO v2.0, with SSRM ownership and optional STRIDE/ATLAS/OWASP/NIST crosswalks.
Layered, evidence-gated security assessments for agentic AI (LLM/RAG/MCP/multi-agent) using MAESTRO v2.0, with SSRM ownership and optional STRIDE/ATLAS/OWASP/NIST crosswalks.
npx skills add Nate-Carroll-Cyber/AI-Threat-Modeling-Claude-SkillLayered, evidence-gated security assessments for agentic AI (LLM/RAG/MCP/multi-agent) using MAESTRO v2.0, with SSRM ownership and optional STRIDE/ATLAS/OWASP/NIST crosswalks.
Production OpenClaw / ClawHub skills by Igor Holt (Genesis Conductor). Persistent state for stateless agents, MCP bridges, thermodynamic verification patterns, and architectural reliability primitives.
An Agent-First CLI for SiYuan (in beta stage now)
Personal AI agent configuration (skills, etc)
Open-source security gateway for LLM APIs — prompt injection detection, PII redaction, dangerous response sanitization, and audit logging. OpenAI/Claude compatible, MCP & Agent SKILL support. Drop-in proxy for AI coding agents (Cursor, Claude Code, Codex).
Claude Code通用技能库
MCP server bridging LLMs to LSP servers for structured code intelligence — definitions, references, call/type hierarchies, hover, symbols, diagnostics. Multi-session shared daemon with refcounted LSP server instances. Starts with C/C++ via clangd, extensible to any language.