AI Agent Hub
Back to plugins
dsh-select-quote preview

dsh-select-quote

Client Updated 2026.09.11

Run the following command in DeepSeek Harness:

dsh plugin install latte03/dsh-select-quote

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

Run dsh plugin install latte03/dsh-select-quote in your terminal to install the plugin; the full source repository is at https://github.com/latte03/dsh-select-quote

About this plugin

In DeepSeek Harness conversations, referencing a passage of context for your next prompt is one of the most common needs. Manual copy-paste mixes quotes into your draft, and hand-writing Markdown markers is slow and error-prone. dsh-select-quote simplifies this into a two-step interaction: select any text in the conversation, a toolbar floats up, and one click drops a quote card into the input. On send, quotes fold into the message by card order, and no Markdown ever appears in the editor.

The plugin provides a text-selection toolbar, stacked multiple quote cards in the input area, draft-free quoting, and card-rendered quotes in the transcript. Your draft stays clean with only your actual question; quotes are injected only at the moment of sending. In the conversation history, the user bubble shows just your input while quotes appear as vertically stacked cards aligned beside it, matching the visual language of the built-in file cards with dark-mode support.

It suits DeepSeek Harness web users who frequently quote conversation context, especially when composing complex prompts from multiple passages and wanting a clean input area free of raw Markdown. Note: pending quotes are held in memory only, so refreshing the page or switching sessions will clear them.

Screenshots

Use Cases

  • Build complex prompts by quickly quoting context passages in multi-turn conversations
  • Arrange multiple text references as clean stacked cards in the input area
  • Skip manual Markdown quote markers to keep the editor tidy

Best For

  • Heavy users of the DeepSeek Harness web client
  • Developers who compose prompts by referencing conversation context
  • Users who want a clean input area free from raw Markdown formatting