Qwen
Installation
SKILL.md
When to Use
User needs Qwen to work reliably for chat, coding, reasoning, structured outputs, or vision. Agent handles surface selection, live model verification, hosted-versus-local tradeoffs, and failure recovery before the workflow reaches production.
Architecture
Memory lives in ~/qwen/. If ~/qwen/ does not exist, run setup.md. See memory-template.md for structure.
~/qwen/
├── memory.md # Status, activation rules, and deployment defaults
├── routes.md # Preferred route per workload
├── servers.md # Known local or hosted endpoints
├── experiments.md # Prompt, parser, and latency notes
└── logs/ # Optional sanitized repro payloads