AI Agent Hub
Back to plugins
🖥️

dsh-selection-toolbar

Client Updated 2026.08.26

Run the following command in DeepSeek Harness:

dsh plugin install suiyideali/dsh-selection-toolbar

Paste the following prompt into your AI chat to install this plugin:

To install this plugin in DeepSeek Harness, run dsh plugin install suiyideali/dsh-selection-toolbar. The source code is hosted at https://github.com/suiyideali/dsh-selection-toolbar.

About this plugin

When using DeepSeek Harness for long task execution or complex workflows, casual queries often clutter the conversation context and interfere with the coherence of subsequent tasks. dsh-selection-toolbar provides one-click text selection interaction, with a toolbar floating above the selection that integrates basic actions like copy, quote, and translate. Its core highlight is the innovative "/btw" feature, inspired by Claude Code, designed to offer a lightweight Q&A method that "doesn't occupy the conversation or interfere with tasks."

When you need a quick explanation for a snippet of code or logs, or when an Agent is running a long task and a sudden question arises, /btw is the perfect choice. It uses an independent routing mechanism to display answers solely within a modal window, completely avoiding writing to chat history or invoking tools. This design ensures the purity of the main conversation while allowing users to obtain contextual assistance anytime, achieving an efficient workflow of "ask and discard."

Whether you are a developer needing frequent context checks or a researcher aiming to keep chat history clean, this plugin significantly boosts productivity. It presents a native-style card and a compact modal window, supports customization, making every selection a precise operation and AI assistance more intuitive.

Use Cases

  • Quickly explain error messages during code review
  • Insert questions anytime while an Agent executes complex tasks
  • Instantly translate or summarize long texts

Best For

  • Developers who frequently need to check context
  • Researchers who want to keep chat history clean
  • Users seeking efficient workflows and instant feedback