superclaude1/dsh-vision-android
C3DeepSeek Harness plugin: multimodal vision (OpenAI-compatible) + Android adb UI automation for real-tap mobile app testing
★ 1+ · superclaude1/dsh-vision-android source on GitHub · this plugin in the registry
superclaude1/dsh-vision-android is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, can rewrite the system prompt, executes system commands, uses eval / new Function.
What it can do
| Capability | Flag | Evidence |
|---|---|---|
| patches the dsh runtime | runtime_patch | ./cordis.patch.yml |
| can rewrite the system prompt | prompt_surface | systemPrompt |
| executes system commands | exec | ×4 src, e.g. lib/index.js:5942, lib/index.js:5945 |
| uses eval / new Function | eval | ×1 src, e.g. lib/index.js:2143 |
| decodes base64 payloads | base64_decode | ×2 src, e.g. lib/index.js:67, lib/index.js:68 |
| reads credential-class env vars | token_env | DSH_VISION_API_KEY |
Services it injects
systemPrompt tools
Outbound domains
api.openai.com
Environment variables it reads
DSH_VISION_BASE_URL DSH_VISION_API_KEY DSH_VISION_MODEL ADB ESBUILD_PATH ANDROID_HOME DSH_MOBILE_SHOTS
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.