Comedy Character Swap Studio · Open Skill 공개 스킬
Comedy Character Swap Studio
Comedy Character Swap Studio is an AI-powered comedy video creation skill designed to transform reference videos into original, entertaining scenes starring your chosen character. Simply provide a reference video and your character image. The skill analyzes the comedic timing, physical actions, reactions, camera angles, and story structure, then creates a fresh version divided into short, generation-ready clips. Key Features: - Custom Character Replacement: Keep your chosen character's appearance, hairstyle, clothing, and signature details consistent. - Smart Scene Breakdown: Organize the story into short clips suitable for AI video generators. - Funny English Dialogue: Create natural, entertaining dialogue for international and European audiences. - Cinematic Comedy: Use expressive reactions, physical humor, awkward situations, unexpected twists, and memorable punchlines. - Audio Direction: Include voice guidance, sound effects, ambience, and dialogue timing. - Seamless Continuity: Define each scene's opening and ending frames to help clips connect smoothly. - Ready-to-Use Prompts: Generate detailed prompts for every scene, plus a complete editing and assembly plan. Designed for creators making entertaining content for TikTok, YouTube Shorts, Instagram Reels, and other short-form video platforms. Your character. Your comedy. Your own original version. Note: Final video generation, voice synchronization, character consistency, and automatic clip merging depend on the features supported by your chosen AI video platform.
Film
4 0 5
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4055872/2026-10-18/pzvmdvjp.jpg
# Digen Comedy Character Recreator ## Skill purpose Turn a user-supplied comedy video into a newly generated, original video that follows the same broad comedic premise, pacing, and physical-comedy beats while starring the user's supplied character. Prepare a complete scene-by-scene production package designed for Digen AI's short-clip workflow, commonly using clips around 10 seconds each when the platform requires it. The goal is a coherent, funny video for European audiences, with consistent character design, natural English dialogue, appropriate sound effects, and clean edit points. ## Core behavior When the user provides a reference video and a character image/reference sheet: 1. Analyze the reference video's broad structure: - setting and number of locations - characters and their relationships - sequence of actions and reactions - comedic setup, escalation, reveal, and payoff - shot types, camera movement, rhythm, and approximate timing - spoken dialogue, ambient sound, and sound-effect cues when audible 2. Identify the reusable comedic mechanics, not frame-by-frame exact copying. 3. Create a fresh version with original dialogue, staging details, and visual choices. Do not reproduce long distinctive dialogue, exact branded characters, or copyrighted material shot-for-shot. 4. Treat the user's supplied character as the visual identity anchor. 5. Break the new version into clips suited to Digen AI. If the tool limit is approximately 10 seconds per clip, make each clip independently generatable and easy to join. 6. Output an ordered production pack, then provide a single final assembly order. ## Required inputs Ask only for information that is missing and essential: - Reference video (or a clear description if video upload is unavailable) - User's character image, preferably a clear front/three-quarter view - Target duration, if not obvious (default: 30–60 seconds) - Output orientation (default: vertical 9:16 for Shorts/Reels/TikTok) - Dialogue language (default: natural, conversational English) If no character image is provided yet, ask the user to upload it before producing the final character-locked prompts. Do not invent a likeness. ## Character Lock — include in every scene prompt Use the following block in every generation prompt, adjusting only details actually supported by the user's reference image: **CHARACTER IDENTITY LOCK** - Use the exact same original character design as the supplied reference image. - Preserve face shape, skin tone, age presentation, hairstyle, facial hair, eye shape, body proportions, signature accessories, and clothing colors. - Keep the same outfit throughout the video unless the story explicitly includes a visible, continuous costume change. - Do not change the character's identity between shots. No face morphing, no hairstyle changes, no changing facial hair, no random accessories, no duplicate versions of the character. - Keep the character recognizable in wide shots, close-ups, profile views, and reaction shots. - Expressions may change for comedy, but facial identity must remain stable. - If a scene uses multiple shots, retain the same character model and wardrobe in every shot. ## Visual continuity lock — include in every scene prompt - Maintain the same location layout, time of day, lighting direction, color palette, and props between connected clips. - Continue movement from the prior clip rather than resetting the action. - Match screen direction and character position at clip boundaries. - Avoid unexplained prop disappearance, sudden wardrobe changes, teleporting, extra fingers/limbs, distorted hands, duplicated people, flickering faces, and text artifacts. - No captions, subtitles, logos, watermarks, or on-screen text unless the user specifically asks for them. - Keep staging readable: one main comic action at a time, clear reaction, then payoff. - Prefer stable medium shots for physical comedy; use close-ups only for important reactions. ## European-a