{"skill":{"slug":"dgx-spark-setup","displayName":"Dgx Spark Setup","summary":"Set up and maintain an NVIDIA DGX Spark (GB10 Blackwell, 128GB unified memory) as a local LLM inference server running vLLM + LiteLLM + OpenClaw. Use when in...","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":107,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1774304761278,"updatedAt":1774306914406},"latestVersion":{"version":"1.0.0","createdAt":1774304761278,"changelog":"- Initial release of DGX Spark setup skill.\n- Provides a step-by-step guide for configuring an NVIDIA DGX Spark (GB10 Blackwell, 128GB unified memory) as a local LLM inference server using vLLM, LiteLLM, and OpenClaw.\n- Covers requirements and instructions for building custom PyTorch and Triton with sm_121 (Blackwell) GPU compatibility.\n- Includes guidance on setting up LiteLLM with per-user virtual keys, running as a systemd user service, and configuring Tailscale for secure remote access.\n- Offers troubleshooting information and pointers for resolving compatibility issues (torch, Triton, flashinfer).","license":"MIT-0"},"metadata":null,"owner":{"handle":"jimmy-hernandez","userId":"s17dy722gamp63azcmjwyx1md983efsw","displayName":"Jimmy Hernandez","image":"https://avatars.githubusercontent.com/u/10274241?v=4"},"moderation":{"isSuspicious":true,"isMalwareBlocked":false,"verdict":"suspicious","reasonCodes":["suspicious.install_untrusted_source","suspicious.vt_suspicious"],"summary":"Detected: suspicious.install_untrusted_source, suspicious.vt_suspicious","engineVersion":"v2.2.0","updatedAt":1774306914406}}