A Prompt Workbench Designed for Video Generation

Brojakhoeman · reddit · 2026-07-16

A Reddit user shared their latest prompt tool project, Prompt-Forge-LD, including a GitHub link and demos. The post highlights numerous features: - Local model support: Ollama, Llama, LM Studio - Scene orchestration: Converts intent into segmented scripts, supporting T2V / I2V, generation/retry/regeneration - Audiovisual and character control: Dialogue density, camera angles, character counts, dual-energy control, music mixing - Complex workflows: Lora stacking, image gallery carousels, frame tools, detail inpainting, theme switching - Local-first: Compatible with local LLMs and ComfyUI The author emphasizes that the project underwent extensive self-generation and self-evaluation iterations, aiming to turn complex generation pipelines into manageable prompt/script tools.

Original post →

More from Multimodal

Multimodal channel →