Skill UI

This skill guides the comprehensive process of crash triage, enabling the assessment of vulnerability exploitability. It covers advanced techniques using debuggers (WinDbg/GDB), sanitizers (ASAN/MSAN), and interpreting low-level data like stack traces and register dumps. It is essential for vulnerability researchers analyzing crash dumps or fuzzer-generated failures to identify root causes and potential attack vectors.

This comprehensive tool guides developers through the entire plugin ecosystem, helping them find and recommend the optimal plugins for any given workflow or task. It covers core coordination, multi-agent collaboration, advanced memory management (Vector DB, RAG), architectural design (ADR/DDD), security auditing, and modern development methodologies. Use it when starting a new project or optimizing existing development pipelines.

Audits GitHub Actions workflows that invoke AI coding agents (Claude Code Action, Gemini CLI, OpenAI Codex, GitHub AI Inference). Detects attack vectors where attacker-controlled input reaches AI agents in CI/CD pipelines, covering trigger events, env var intermediaries, tool allowlists and sandbox configuration.