Analyze source video and swap the on-camera subject via the ralphy pipeline.
npx skills add https://github.com/alecs5am/ralphy --skill ugc-model-swap
Please help me install this Agent Skill. Skill: ugc-model-swap Source: https://github.com/alecs5am/ralphy/tree/main/.agents/skills/ugc-model-swap Command: npx skills add https://github.com/alecs5am/ralphy --skill ugc-model-swap
The craft specialist for the swap kind of remix. The user provides a source clip and a replacement person/character; this skill mirrors the source faithfully and changes only the on-camera subject. It enables rapid creation of multiple variants without re-rigging the entire scene.
Provide a source clip and a replacement character, then run the single-continuous-scene prompt workflow.
High-intent search queries and answers about installing and using this skill.
To swap an on-camera actor while keeping the background intact, you provide a source clip and a replacement character to generate a controlled subject replacement that preserves all original scene elements.
The best way to do a face swap on UGC video ads is using a single-continuous-scene prompt that describes the entire video as one flowing paragraph, ensuring face-lock and composition rules maintain realism.
To keep a replacement character's face centered during a video swap, apply face-lock and composition rules within your prompt structure to ensure the subject remains the focal point of the frame.
You can generate video remixes of named real people, but the workflow enforces reference gating and strict safety policies when a replacement involves a real person to ensure compliant generation.
Your AI video subject replacement loses track of props because it lacks prop and body-mechanics discipline, requiring explicit measurable movements and precise interaction descriptions in the prompt.
Yes, you need a master reference image for replacement to accurately mirror the source clip and swap only the on-camera subject while driving the scene through the generation pipeline.
fal-ai-communityGenerate consistent character designs and reference sheets from a character anchor.
devMoezGenerate Manim Community Edition animations from text prompts into rendered video.
team-attentionGenerate structured markdown cheatsheet prompts from long-form content.
nolanx-aiGenerate image-to-video prompts with identity, style, and camera continuity.
