seedance-prompt-review-skill

Audit Seedance 2.0 dynamic prompts for fidelity, execution, and asset-reference issues.

11|10|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/ccnuzw/Seedance-anti --skill seedance-prompt-review-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: seedance-prompt-review-skill
Source: https://github.com/ccnuzw/Seedance-anti/tree/main/feicai-studio/resources/builtin-skills/seedance-prompt-review-skill
Command: npx skills add https://github.com/ccnuzw/Seedance-anti --skill seedance-prompt-review-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, repeatable audit process to evaluate Seedance 2.0 dynamic prompts produced by storyboard artists, ensuring prompts faithfully implement the director's intent, are Seedance-executable, and reference assets correctly.

Core Features & Use Cases

  • Step-by-step audit checklist that compares each prompt line-by-line with the director's scene analysis and character/scene asset files.
  • Mental simulation guidance and a multi-dimension scoring rubric (fidelity, visual restoration, action executability, camera feasibility, Seedance-friendliness, audio design, emotional accuracy).
  • Asset mapping and platform-constraint checks that verify @references, image/video/audio limits, and scene reference numbering.
  • Output templates for per-prompt scoring tables, average score calculation, PASS/FAIL decision rules, and prioritized issue reports.
  • Use case: A director reviews phase-three prompt outputs (outputs/<episode>/02-seedance-prompts.md) to produce a remediation list and final acceptance stamp before generation.

Quick Start

Review the episode's 02-seedance-prompts.md against 01-director-analysis.md and assets/character-prompts.md to produce the scoring table, average score, PASS/FAIL decision, and a prioritized issue list.

Frequently Asked Questions about seedance-prompt-review-skill

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

FAQPage Schema
How do I audit Seedance 2.0 dynamic prompts for director-aligned video production?

To audit Seedance 2.0 dynamic prompts, perform a line-by-line comparison against director analysis and asset files, mentally simulate generated frames, and apply a multi-dimension scoring rubric to evaluate fidelity and execution. This produces a prioritized remediation list and a Pass/Fail decision.

What is a prompt review checklist for episodic video storyboard outputs?

A prompt review checklist for storyboard outputs verifies fidelity, visual restoration, action executability, camera feasibility, Seedance-friendliness, audio design, and emotional accuracy. It ensures phase-three prompts faithfully implement the director's scene intent and reference assets correctly before generation.

How do I check asset mapping and platform constraints in Seedance prompts?

To check asset mapping in Seedance prompts, verify @references, image/video/audio limits, and scene reference numbering. This platform-constraint check ensures prompts are Seedance-executable and correctly map to the character and scene asset files.

What is the best way to evaluate prompt fidelity and action executability for video generation?

The best way to evaluate prompt fidelity and action executability is using a multi-dimension scoring rubric alongside mental simulation of generated frames. This identifies execution issues and verifies that dynamic prompts align with the director's scene analysis.

Can I use this prompt audit process for phase-three character and scene prompts?

Yes, you can use this audit process for phase-three storyboard prompt outputs, character prompts, and scene prompts. It compares these outputs against the director analysis and asset files to produce a final acceptance stamp before generation.

Why does my Seedance prompt fail the director acceptance review?

A Seedance prompt fails the director acceptance review if it has fidelity, execution, or asset-reference issues. The audit identifies these failures through line-by-line comparison and multi-dimension scoring, outputting a prioritized remediation guidance list.