seedance-2-5-video-director

Generates director briefs, scripts, and copy-ready prompts for Dreamina Seedance 2.5 video generation.

348|46|Updated Aug 18, 2026
One-click install
npx skills add https://github.com/liyue-aigc/seedance-2-5-video-director --skill seedance-2-5-video-director-liyue-aigc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: seedance-2-5-video-director
Source: https://github.com/liyue-aigc/seedance-2-5-video-director
Command: npx skills add https://github.com/liyue-aigc/seedance-2-5-video-director --skill seedance-2-5-video-director-liyue-aigc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Writing effective prompts for Dreamina/即梦 Seedance 2.5 requires knowing its modes, duration limits, asset-reference syntax, and camera grammar. This Skill turns raw video ideas, reference images, videos, audio, storyboards, or old prompts into structured director plans, diagnoses, and copy-ready Seedance 2.5 prompts without submitting any paid generation job. ## Core Features & Use Cases - Eight primary modes: basic 4-30s multimodal video, exact 30-second timelines, 30-180s Long Video, video extension (prepend/append), video editing, Clay Renderer white-model rendering, seamless two-video transitions, and multi-grid storyboard animation. - Specialist capabilities: character and wardrobe locking, multi-person binding, timbre reference, BGM removal, creative transfer, perspective reconstruction, green-screen compositing, and realistic acted dialogue with physical causality. - Prompt diagnosis and revision: detect contradictions in timing, camera, character, physics, and audio in existing prompts, or rewrite them while preserving the original structure. - Use Case: Provide a character reference image and a 20-second water-obstacle-course story; receive a full director plan with an asset lock table, continuity rules, and one final copy-ready Seedance 2.5 prompt. ## Quick Start Use $seedance-2-5-video-director to turn my video idea, reference images, and desired duration into a complete director plan and final Seedance 2.5 prompt.

Frequently Asked Questions about seedance-2-5-video-director

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a prompt for Seedance 2.5 video generation?▼

Invoke the Skill with your task, duration, aspect ratio, asset roles, story beats, and sound policy. It selects one primary mode, builds an asset lock table, and returns a director plan plus one copy-ready Seedance 2.5 prompt.

How do I extend an existing video with Seedance 2.5?▼

State whether the new interval goes before or after the original clip and its duration. The Skill normalizes direction to prepend-before-source or append-after-source, describes only the new interval, and protects the original footage.

Does a person reference image in Seedance 2.5 only copy the face?▼

No. By default a person reference locks the complete visible character: identity, face, hair, body proportions, wardrobe, footwear, and accessories. Only an explicit 'face only' instruction narrows the scope.

What are the input limits for Seedance 2.5 reference media?▼

Up to 30 images, 10 reference videos totaling 30 seconds, and 10 audio clips totaling 30 seconds per request. Videos must be MP4/MOV at 24-60 FPS and 2-30 seconds each; the Skill validates these limits before drafting.

Can this Skill generate the video directly or spend Dreamina credits?▼

No. It produces only planning text, scripts, diagnoses, and prompts. It never submits generation jobs, consumes credits, or claims a video was created; you paste the final prompt into Dreamina yourself.

Why does my Seedance prompt produce conflicting camera or physics results?▼

Conflicts like one-take versus shot/reverse-shot, fixed distance versus face-filling close-ups, or reactions appearing before their cause cause unstable output. Use diagnosis-only mode to identify contradictions before rewriting.