What is prompt-forge-ld?

PromptForge LD is a custom node pack for ComfyUI, the visual tool many people use to build AI image and video generation workflows. It focuses on writing shot scripts for LTX Video, a model used to turn images or text into short video clips. Instead of typing one line of instructions into a plain text box, this pack gives you a full panel for writing motion scripts: body movement the model can actually render, dialogue that stays readable, continuity between clips, and optional accents or styles you can turn on when you want them. The goal stated by the author is simple: better generated scripts, fewer broken frames, and less back and forth fighting with the model to get a usable result. The pack has three parts. PromptForge LD is the main writing panel where you generate or refine a script. PromptForge Unpack takes that result and splits it into the separate pieces a video workflow needs: image, positive prompt, negative prompt, width, and height. LoraForge LD is a third node for stacking multiple LoRA files, which are small add-on files that adjust how a model behaves, with separate strength controls for video and audio. To use it, you install the node into ComfyUI's custom_nodes folder and restart. The node itself does not include any AI model. You need to run a language model separately, either through LM Studio, a beginner-friendly app with its own interface, or through llama.cpp's llama-server for people who want a standalone server with more control. Once that local model server is running, you point PromptForge LD at its address and paste in the model's name so the two can talk to each other. The author is upfront that this is a personal project shared publicly, built for their own use first. The README stays family-friendly, though the node itself is described as a general-purpose script writer where the subject matter is left up to the person using it.