A local AI desktop workspace built on DeepSeek Harness, bringing sessions, projects, files, web research, plugins, and Office artifacts together in one app for daily work.
dsh-shell
Deep Shell is an intelligent terminal application that combines a chat interface with a command-line interface, It helps users execute terminal commands through natural language conversations.
Plugin facts
Screenshots
About this plugin
Deep Shell is an intelligent terminal application that combines a chat interface with a command-line interface, It helps users execute terminal commands through natural language conversations.
Install this DSH plugin
Copy the command below and run it in your DeepSeek Harness terminal.
dsh plugin add github:gokul6350/dsh-shell ⚠️ Plugins run with the permissions of your dsh process and may execute code at install time. Review the source repository and its license before installing. Pin the commit hash for reproducible installs. Plugin safety →
More about this plugin
Deep Shell 🐚 Note: For the Linux version, see @gokul6350/DeepShell4 Deep Shell is an intelligent terminal application that combines a chat interface with a command-line interface, powered by Google's Gemini AI.
About this plugin: dsh-shell
What is dsh-shell?
dsh-shell is a sessions & messages plugin maintained by gokul6350. Deep Shell is an intelligent terminal application that combines a chat interface with a command-line interface, It helps users execute terminal commands through natural language conversations.
How do I install dsh-shell?
Run dsh plugin add github:gokul6350/dsh-shell in your DeepSeek Harness terminal. The dsh CLI resolves the plugin from GitHub and installs it into your active profile. For reproducible installs, pin a commit hash: dsh plugin add github:gokul6350/dsh-shell#commit.
Is dsh-shell free?
dsh-shell is a community open-source project released under the MIT license. You can inspect its source and install it for free.
Related plugins
dsh-router-standard
⭐ 57A task-aware reasoning-mode router for DeepSeek Harness. Each task is placed on one continuous react-to-spec axis, the router picks from three measured behavior bands before the first model request, and persona plus first-turn tool injection are supported with agent-visible tuning.
Rewind conversation and workspace state in DeepSeek Harness — undo agent turns and restore code state when things go sideways.