AI Agent Hub
Back to plugins
🖥️

dsh-message-injector

Client Updated 2026.08.30

Run the following command in DeepSeek Harness:

dsh plugin install wf-ping/dsh-message-injector

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

Run dsh plugin install wf-ping/dsh-message-injector in your DeepSeek Harness terminal to install; the project repository is available at https://github.com/wf-ping/dsh-message-injector .

About this plugin

When chatting with dsh (DeepSeek Harness), many users start every message with a fixed prefix: a skill invocation, a language directive, or a format requirement. Retyping it each turn is tedious and easy to forget, and one slip can derail the entire session.

dsh-message-injector turns those repetitive prefixes into preset groups. Create them in the plugin config (plain text or /-prefixed skill calls), select one, and every time the input box is empty the content is auto-filled within about 500 ms. If the box already holds text the plugin stays quiet. Missing skills trigger a config-card warning and are skipped at injection time so nothing silently misfires. Selection state is persisted to $DSH_HOME/settings.yaml across sessions, reloads, and tabs.

If your DeepSeek workflow relies on a consistent prompt prefix or skill combo, this plugin lets you focus on the question rather than on reformatting every message.

Use Cases

  • Manually typing the same skill invocation prefix or format directive on every message
  • Switching between multiple prompt template groups quickly
  • Eliminating repetitive manual prefix entry that leads to omissions or format drift

Best For

  • dsh users who drive DeepSeek workflows with fixed prompt prefixes
  • Developers who frequently chain multiple skill calls and need preset prefixes
  • Users who want to cut down repetitive typing and speed up each conversation turn