RichDavidMu/create-dsh-plugin
C2Scaffold a DeepSeek Harness plugin project — a working plugin with one model-facing tool, a profile bundle that mounts it, dsh's own toolchain, and documentation an agent can follow without reading ds
★ 0 · RichDavidMu/create-dsh-plugin source on GitHub · this plugin in the registry
RichDavidMu/create-dsh-plugin is a DeepSeek Harness plugin rated C2 — one powerful capability or sensitive behavior. It runs code at install time, executes system commands, powerful capability in test/example code only.
A skill, not an installable plugin — ships SKILL.md with no plugin manifest
What it can do
| Capability | Flag | Evidence |
|---|---|---|
| runs code at install time | install_script | postinstall: node scripts/install-lefthook.mjs |
| executes system commands | exec | ×27 in authored code, e.g. scripts/install-lefthook.mjs:9, scripts/install-lefthook.mjs:9 |
| powerful capability in test/example code only | dev_surface | in code that does not ship: systemPrompt |
Environment variables it reads
DSH_GRAPH
How to read this
Levels measure capability surface and transparency, not maliciousness. A C3 plugin can be entirely legitimate — a desktop shell genuinely needs subprocesses. The point is that you can see this before installing. See the levels explained and how dsh plugins work.
Findings come from static analysis of shipped code; nothing is executed. Think a flag is wrong? Open an issue — every flag cites the file and line it came from.