Skills LibraryTheme

Skills Library / Customer Response Drafter / for Local Models

Customer Response Drafter — for Local Models

This is the Local Models version of the Customer Response Drafter skill: draft support replies that resolve: acknowledge, answer, next step, in the right tone. For local models (Llama, Mistral, GLM via Ollama or LM Studio) the skill becomes the system prompt: the text below goes in the SYSTEM block of a Modelfile or the system-prompt field of your runner, saved as a named specialist.

The file

✎ Editable — click in, customize or extend this prompt, then copy. Your edits copy with it.Take this prompt into the Generator and customize it to your own specification →
Download system-prompt.txt

Install

  1. Put the text in the SYSTEM block of a Modelfile (or the system field of your runner)
  2. Set temperature low (0.2-0.4) - procedures want consistency, not variety
  3. Build the named model and test it on one real input before relying on it

Other platforms

Claude versionChatGPT versionGemini versionMicrosoft Copilot versionQwen & Kimi version

Canonical version and the skill's full description: Customer Response Drafter. The underlying procedure is identical on every platform - only the packaging differs, because each model takes its skills its own way.

AD PLACEMENT — FOOTER BANNER

Mpromp2 — the AI Prompt Generator. Every guide and skill file here is free to use, including commercially.