Sign In

Civitai Auto-Trigger Append (Native Conditioning Node for ComfyUI)

Updated: May 19, 2026

tool

Download

1 variant available

Archive Other

4.95 KB

Verified:

Type

Other

Stats

4

Reviews

Published

May 19, 2026

Base Model

SDXL 1.0

Hash

AutoV2
D78A384734

Stop Copy-Pasting LoRA Trigger Words!

Tired of alt-tabbing back to Civitai to find the exact activation tags for a LoRA you downloaded last week? This lightweight, zero-dependency custom node automates the entire process directly on your ComfyUI canvas.

It reads your local .civitai.info metadata files on the fly, extracts the official Trained Words specified by the creator, combines them with your base prompt text, and natively encodes everything into CONDITIONING data ready for your KSampler.

Key Features

  • Native Conditioning Output: Replaces the standard CLIP Text Encode node entirely. It tokenizes and encodes your prompt with proper SDXL/Flux pooled output tensors seamlessly.

  • Dynamic Prompt Mixing (3 Modes):

    • Append to End: Keeps your core prompt first, adding Civitai tags at the end.

    • Prepend to Start: Forces trigger words to the absolute front of the prompt string for maximum mathematical weight in the sampler (Perfect for heavy style/morph LoRAs!).

    • Only Output Triggers: Isolates and extracts the raw tags to pass downstream.

  • Fail-Safe Design: If a LoRA is missing its .civitai.info metadata file, the node gracefully passes your base text through without throwing errors or breaking your generation queue

Quick Connection Guide

  1. Route your Checkpoint/LoRA CLIP output (yellow dot) straight into the clip input of this node.

  2. Type your core descriptive prompt into the node's text box.

  3. Select your target LoRA from the dropdown.

  4. Connect the orange conditioning output dot directly into the positive input of your KSampler.

(Tip: Set the action dropdown to Prepend to Start if you want the LoRA's style to hit with maximum intensity! See the car preview images for a side-by-side comparison of Append vs Prepend.)