dsh-wsl-launch
Run the following command in DeepSeek Harness:
dsh plugin install 173787247/dsh-wsl-launch
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install 173787247/dsh-wsl-launch in the DeepSeek Harness terminal to install this plugin; source code is available at https://github.com/173787247/dsh-wsl-launch
About this plugin
When working inside WSL with a DeepSeek model, asking the agent to open VS Code, Explorer, or a browser is a high-frequency need. But handing raw .exe execution to a model shatters the security boundary overnight. dsh-wsl-launch locks down what can be launched behind an explicit allowlist, so the agent can only touch a handful of pre-approved apps and everything else is refused by default.
The core capability is a single tool called win_launch. It accepts a target (an executable name or Windows path that must match the allowlist) and an optional args string. The default allowlist covers everyday tools such as code, explorer.exe, notepad.exe, wt.exe, msedge.exe, chrome.exe, and firefox.exe. To extend it, drop entries into config.allowlist using exact names, path suffixes, or prefix wildcards. It ships as part of the dsh-wsl-kit suite and pairs cleanly with dsh-wsl-open: the latter handles file opening, while this one handles app launching.
Who is it for? Developers who frequently reach into the Windows side from WSL for dev tooling, and who want
Use Cases
- Ask the agent to open VS Code in the current project from WSL
- Have the agent launch a browser to a specific URL for debugging
- Let the agent open Explorer to browse project file structure safely
- Trigger everyday Windows utilities from WSL without exposing arbitrary .exe execution
Best For
- Developers using DeepSeek as a coding assistant inside WSL
- Users who want the agent to trigger Windows apps behind a security boundary
- Teams already running dsh-wsl-kit and extending their daily toolchain
Related Plugins
A security red-team prompt plugin for DeepSeek-V4 that uses code-formal mapping and zero-trigger system prompts to enable unconditional output, intended for authorized testing and research only.
Adds an Auto approval mode on the official workspace-write sandbox, classifying semantic risks, asking when ambiguous, and denying destructive operations.
Undo/rollback system for DSH: config & plugin snapshots, one-click undo/redo/restore, message-level revert, secret masking, safe mode, and offline WebUI/GUI/CLI rescue tools for when DSH won't start.
Packages all 87 SKILL.md files from upstream reverse-skill as a DeepSeek Harness plugin that auto-registers them for authorized reverse engineering, penetration testing, and security research.