seedance-debugger SKILL.md skillA

seedance-debugger from cclank/lanshu-awesome-ai-video-kit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add cclank/lanshu-awesome-ai-video-kit/seedance-debugger@git:20260526.b5da140

Pins git:20260526.b5da140 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
cclank/lanshu-awesome-ai-video-kit
Path
skills/seedance-debugger/SKILL.md
Versions recorded
3
Latest version
git:20260526.b5da140, committed May 26, 2026
First version recorded
May 25, 2026
Size
12,764 bytes
Repository stars
429
License
MIT

诊断 Seedance 2.0 生成视频时出现的常见问题(人物 ID 漂移/双胞胎/字幕/Logo/风格漂移/延长跳变/画质劣化/特效不对/中文发音/音色不准/结尾噪音 等 12 类),定位根因并给出修复后的提示词。用于"我的提示词生成出来不对"、"视频里出现奇怪的字幕"、"人脸不像参考图"、"出现两个一样的人物"、"风格变了"、"怎么修这个提示词"等触发场景。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# seedance-debugger

输入:有问题的提示词 + 用户描述的症状(或视频截图)。
输出:诊断(根因) + 修复后的提示词 + 解释为什么这样改。

## 何时不用此 skill

- 提示词完全没写,只是想要新提示词 → [seedance-prompter](../seedance-prompter/)
- 想拆复杂剧情 → [seedance-storyboard](../seedance-storyboard/)

## 12 类问题诊断手册

详见 [methodology/08-避坑12问.md](../../methodology/08-避坑12问.md)。这里给出快速决策树。

### 症状识别 → 跳转

| 用户说的症状 | 跳转 |
|---|---|
| "人脸不像 / 中途换脸 / 撞脸明星" | [#1 人物-id-漂移](#1-人物-id-漂移) |
| "莫名出现字幕 / 视频里有文字" | [#2 字幕](#2-字幕) |
| "有水印 / 有 Logo" | [#3 logo-水印](#3-logo-水印) |
| "动漫风格变成真人了 / 风格变了" | [#4 风格漂移](#4-风格漂移) |
| "拼接处跳帧 / 衔接不流畅" | [#5 延长跳变](#5-延长跳变) |
| "出现两个一模一样的人" | [#6 双胞胎](#6-双胞胎) |
| "延长后画质变糊 / 人脸有斑驳色块" | [#7 画质劣化](#7-画质劣化) |
| "特效不对 / 倒计时乱跳" | [#8 特效不对](#8-特效不对) |
| "人物多了/少了" | [#9 参考人物过多](#9-参考人物过多) |
| "结尾有咔哒声" | [#10 结尾噪音](#10-结尾噪音) |
| "中文读错字了" | [#11 中文发音](#11-中文发音) |
| "音色不像参考音频" | [#12 音色不准](#12-音色不准) |

---

### 1. 人物 ID 漂移

**根因**:参考图混用(人脸 + 全身 + 服装在一张图) OR 人脸占比过小。

**修复策略**:
1. 让用户**额外准备一张大头照**(只包含面部,无表情最佳)
2. 提示词中明确区分:`<主体1> 的面部特征参考图片1(大头照),妆造参考图片2(全身照)`
3. **重要素材前置**

**反例**:`将图1中的人物作为主角...`
**正例**:`将@图片1(大头照)的面部特征作为主角面部锚点,@图片2(全身照)作为整体妆造参考。主角...`

---

### 2. 字幕

**根因**:参考图/视频中包含文字;竖屏更易触发。

**修复策略**:
1. 加约束:`保持无字幕,避免生成任何文字或字幕`
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add cclank/lanshu-awesome-ai-video-kit/seedance-debugger@sha256:bf0d5ed0bbe0163b

The line at the top of this page pins the label git:20260526.b5da140, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_vw5mxwv72ja2pmpn.svg)](https://markdownregistry.com/a/art_vw5mxwv72ja2pmpn)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 3 versions of this file, between May 25, 2026 and May 26, 2026, more than one a day.

versioncommittedcommitsizeaudit
git:20260526.b5da140 latest2026-05-26 b5da140 12,764 BA view · diff
git:20260526.861facc2026-05-26 861facc 12,521 BA view · diff
git:20260525.bb3aa492026-05-25 bb3aa49 7,171 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (12764 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

cclank/lanshu-awesome-ai-video-kit · 429 stars · license MIT · pushed 2026-10-07 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_vw5mxwv72ja2pmpn
GET https://markdownregistry.com/api/v1/resolve?ref=cclank/lanshu-awesome-ai-video-kit/seedance-debugger
GET https://markdownregistry.com/api/v1/blob/bf0d5ed0bbe0163b604e3f10d43dbaaa8775ac3b2b8f2f6cab8c3769a453891b

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from cclank/lanshu-awesome-ai-video-kit

happyhorse-prompter SKILL.md skill
cclank/lanshu-awesome-ai-video-kit · skills/happyhorse-prompter/SKILL.md · 生成符合 HappyHorse 1.0 严格规则的紧凑提示词(30-55 词),主体先行 + 明确镜头技术 + 音频激活路径(with X audible / speaking English at natural…
git:20260526.2f01e20 · audit A · 429 stars
kling-prompter SKILL.md skill
cclank/lanshu-awesome-ai-video-kit · skills/kling-prompter/SKILL.md · 生成符合 Kling 3.0(可灵 3.0,快手)规则的视频提示词。三种写法自适应:4 部分基础公式(短视频)/ 5 层进阶公式(剧情+音频)/ 图生视频专用(只描述运动)。Kling 是 2026 年中文理解最强、原生音画同步、最长 2…
git:20260526.2f01e20 · audit A · 429 stars
model-selector SKILL.md skill
cclank/lanshu-awesome-ai-video-kit · skills/model-selector/SKILL.md · 根据用户的视频需求(场景、时长、音频、语言、平台限制、预算、是否需要本地部署/角色一致性等),从 16 个主流 AI 视频模型(12 商业 + 4 开源)中推荐最匹配的 1-3 个,并解释为什么。覆盖商业:Seedance…
git:20260527.196eed7 · audit A · 429 stars
prompt-translator SKILL.md skill
cclank/lanshu-awesome-ai-video-kit · skills/prompt-translator/SKILL.md · 把一条 AI 视频提示词从源模型(如 Sora 2)的写法风格转换为目标模型(如 Kling 3.0 / Wan 2.7 / Veo 3.1 等)的最佳实践写法。基于 110 条 10 场景 × 11…
git:20260526.861facc · audit A · 429 stars
seedance-prompter SKILL.md skill
cclank/lanshu-awesome-ai-video-kit · skills/seedance-prompter/SKILL.md · 把用户的自然语言视频需求转换为符合 Doubao Seedance 2.0 进阶公式的提示词(8 要素:精准主体+动作细节+场景环境+光影色调+镜头运镜+视觉风格+画质+约束条件)。用于"帮我写一个 Seedance…
git:20260530.a8bba67 · audit A · 429 stars
seedance-storyboard SKILL.md skill
cclank/lanshu-awesome-ai-video-kit · skills/seedance-storyboard/SKILL.md · 把复杂剧情/故事大纲拆分为 Seedance 2.0 的"镜头1/镜头2/镜头3"多分镜结构。每个分镜按 4 维度组织(运镜 + 主体动作与表情 + 位置/空间变化 + 音频)。用于"帮我把这个故事写成 Seedance…
git:20260526.2f01e20 · audit A · 429 stars

Every file in cclank/lanshu-awesome-ai-video-kit

Browse by kind, by grade A, or by owner.