vibeinging/deepseek-harness-desktop-app
C3DeepSeek Harness Desktop App: a local AI desktop workspace for DSH Sessions, projects, files, web research, plugins, and Office artifacts.
★ 100+ · vibeinging/deepseek-harness-desktop-app source on GitHub · this plugin in the registry
vibeinging/deepseek-harness-desktop-app is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, runs code at install time, can rewrite the system prompt, gates tool execution.
What it can do
| Capability | Flag | Evidence |
|---|---|---|
| patches the dsh runtime | runtime_patch | ./cordis.patch.yml |
| runs code at install time | install_script | postinstall: patch-package |
| can rewrite the system prompt | prompt_surface | webServer, system-prompt/assemble, tools/pre-execute |
| gates tool execution | tool_gate | hook: tools/pre-execute |
| executes system commands | exec | ×54 src, e.g. electron/artifact-context-menu.js:3, electron/main.js:16 |
| decodes base64 payloads | base64_decode | ×14 src, e.g. electron/main.js:681, electron/main.js:1419 |
| starts a network server | net_server | ×24 src, e.g. eval/browser-workspace-ui-smoke.mjs:14, eval/browser-workspace-ui-smoke.mjs:26 |
Services it injects
agents locale sessions slots theme tools webServer
Hooks it attaches
agent/created agent/disposed agent/pre-step agent/request system-prompt/assemble theme/change tools/pre-execute
Outbound domains
api.openai.com api.example.com open.feishu.cn duckduckgo.com search.example.com api.deepseek.com
Environment variables it reads
DSH_USER_DATA_DIR DSH_EVAL_HOME DSH_EVAL_ISOLATED DSH_EVAL_MODE DSH_NODE_BIN DSH_EVAL_DB_SQLITE_PATH DSH_DATA_ROOT DSH_EVAL_SITE_EXPORT_PATH DSH_EVAL_VERBOSE DSH_RUNTIME_DISTRIBUTION
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.