Diagnoses why ROCm, the HIP SDK, PyTorch, or llama.cpp is broken on an AMD GPU on Linux, Windows, or WSL2, then applies a low-risk fix with consent or hands back the exact next step. Also routes Lemonade, LM Studio, and Ollama problems to the right upstream channel.
安装
npx skills add amd/skills --skill rocm-doctornpx skills add amd/skills --skill rocm-doctor -a claude-code -gnpx skills add amd/skills --skill rocm-doctor -a codex -ggemini skills install https://github.com/amd/skills.git --path skills/rocm-doctorgh skill install amd/skills rocm-doctorgit clone --depth 1 https://github.com/amd/skills /tmp/amd__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/amd__skills/skills/rocm-doctor ~/.claude/skills/rocm-doctor # Claude Code
cp -R /tmp/amd__skills/skills/rocm-doctor ~/.agents/skills/rocm-doctor # Codex, Cursor, Gemini CLI, Copilot, OpenCode