
Seedance Debugger
This is a forensic debugger for when your Seedance 2.0 video generation goes wrong. You paste in your broken prompt and describe what happened (duplicate characters, random subtitles, face drift, style switching mid-video), and it walks through a decision tree of 12 common failure modes to diagnose root cause. Then it hands back a fixed prompt with a diff table explaining each change. The methodology is extremely specific: it knows that three-view reference images cause twin cloning, that café scenes trigger storefront text, that extending videos more than a few times degrades faces into blotchy messes. Most fixes are prompt rewrites, but for things like splice jumpcuts it tells you to drop 6 frames in one clip and 1 in the other because that's the frame offset issue. Honest and procedural.
- GitHub stars
- 269
- last updated
- Jun 1, 2026
Install
$ npx -y skills add cclank/lanshu-awesome-ai-video-kit --skill seedance-debugger --agent claude-codeInstalls into .claude/skills of the current project.
Files
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
- 想拆复杂剧情 → seedance-storyboard
12 类问题诊断手册
详见 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 |
| "莫名出现字幕 / 视频里有文字" | #2 字幕 |
| "有水印 / 有 Logo" | #3 logo-水印 |
| "动漫风格变成真人了 / 风格变了" | #4 风格漂移 |
| "拼接处跳帧 / 衔接不流畅" | #5 延长跳变 |
| "出现两个一模一样的人" | #6 双胞胎 |
| "延长后画质变糊 / 人脸有斑驳色块" | #7 画质劣化 |
| "特效不对 / 倒计时乱跳" | #8 特效不对 |
| "人物多了/少了" | #9 参考人物过多 |
| "结尾有咔哒声" | #10 结尾噪音 |
| "中文读错字了" | #11 中文发音 |
| "音色不像参考音频" | #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:
- 让用户额外准备一张大头照(只包含面部,无表情最佳)
- 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) - 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。
Sponsors
Details
- Repository
- github.com/cclank/lanshu-awesome-ai-video-kit
- Categories
- AI & Agent Building, Debugging
- License
- MIT
- First seen
- Jun 11, 2026