dsh-wsl-docker
Install set: part of dsh-wsl-kit. Prefer KIT_SET=daily | llm | github | full (see kit README). Fault tree: TROUBLESHOOTING.md.
DeepSeek Harness plugin: Docker Desktop vs WSL engine doctor, plus vLLM / OpenAI :8000 container hints.
中文说明 → README.zh.md
Where it sits
Checks Docker Desktop versus the WSL context, and whether a local inference port is actually an API.
flowchart LR
agent["dsh agent"] --> tool["docker_doctor"] --> docker["Docker context and local API"]
Suite diagram and version snapshot: dsh-wsl-kit. This plugin is 0.2.2 (full; also in llm). Do not copy that matrix into this README.
Compatibility
| Field | Value |
|---|
| Plugin | dsh-wsl-docker 0.2.2 |
| Minimum dsh | ≥ 0.1.2 (web UI one-shot ?token= on Windows relay :3081) |
| Latest verified | See dsh-wsl-kit Compatibility (currently 0.1.7-alpha.2) — single source of truth for the suite |
| Kit set | llm / full (some also useful alone) |
| Cloud Flash | Use model id deepseek-flash (V4.1 Flash) in ~/.dsh/settings.yaml / llm-deepseek — not configured by this plugin |
| Agent Teams | Upstream experimental; not required here |
Suite floor versions: kit check-plugin-versions.sh. Fault tree: TROUBLESHOOTING.md.
What it does
Tool docker_doctor (focus: all | daemon | vllm):
- Reports CLI path, context, server version, nvidia runtime hint
- Lists running containers;
focus=vllm filters OpenAI/:8000-ish names
- Advises pairing with
host_reach for baseURL
Install
Included in KIT_SET=llm and full:
dsh plugin --profile web add github:173787247/dsh-wsl-docker
Restart dsh web; open a new session.
License
MIT
