muapi-storyboard
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseStoryboard Generator
故事板生成器
Generate N keyframes for a short story or scene sequence (image only, no video).
Estimated credits: ~24 per run.
为短篇故事或场景序列生成N个关键帧(仅生成图片,不生成视频)。
预估积分消耗: 每次运行约24积分。
Inputs
输入参数
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
| text | yes | — | One-line story premise (e.g. "lonely robot finds a tiny mechanical bird friend"). |
| int | no | 6 | Number of keyframes to produce. |
| text | no | cinematic, photoreal, soft lighting, 16:9 | Visual style tags applied to every keyframe. |
| 名称 | 类型 | 是否必填 | 默认值 | 描述 |
|---|---|---|---|---|
| 文本 | 是 | — | 一行式故事前提(例如:“孤独的机器人找到了一个小小的机械鸟朋友”)。 |
| 整数 | 否 | 6 | 要生成的关键帧数量。 |
| 文本 | 否 | cinematic, photoreal, soft lighting, 16:9 | 应用于每个关键帧的视觉风格标签。 |
Steps
步骤
Use the plan to dispatch all N keyframes in a single parallel layer.
- Decompose into
premisestory beats with a clear arc: setup → inciting moment → escalation → climax → resolution.{{scenes}}- Each beat gets a one-paragraph visual description.
- Maintain character / object continuity across beats (same character appearance, same world).
- For each beat, create a node (model=nano-banana-2, aspect_ratio=16:9):
muapi image generate- Prompt = .
"<beat description>. {{style}}" - Tier: balanced (these are reference keyframes, not finals).
- Aspect ratio: 16:9.
- Prompt =
- Run the plan in parallel (no between keyframes).
depends_on - Return the asset ids in beat order with a one-line caption per scene.
使用规划在单个并行层中调度所有N个关键帧。
- 将分解为
premise个具有清晰叙事弧的故事节拍: 铺垫 → 触发事件 → 升级 → 高潮 → 结局。{{scenes}}- 每个节拍对应一段视觉描述。
- 在各个节拍间保持角色/物体的连贯性(相同的角色外观、相同的世界设定)。
- 为每个节拍创建一个节点(model=nano-banana-2, aspect_ratio=16:9):
muapi image generate- 提示词 = 。
"<beat description>. {{style}}" - 层级:balanced(这些是参考关键帧,而非最终成品)。
- 宽高比:16:9。
- 提示词 =
- 并行执行规划(关键帧之间无依赖)。
depends_on - 按节拍顺序返回资产ID,并为每个场景添加一行说明文字。
Notes
注意事项
- Don't animate, upscale, or add audio — this skill is keyframes only.
If the user wants video, suggest the skill afterward.
music-video - For consistency, repeat character description verbatim in every prompt ("a small rusty humanoid robot with…") rather than relying on the model to remember.
- 不要进行动画制作、放大或添加音频——本技能仅生成关键帧。如果用户需要视频,后续可推荐技能。
music-video - 为保证一致性,在每个提示词中逐字重复角色描述(例如:“一个小型生锈类人机器人,拥有……”),而非依赖模型记忆。
Trigger Keywords
触发关键词
storyboardkeyframesscene sequencestory panelsstoryboardkeyframesscene sequencestory panelsNotes for the Executing Agent
执行Agent注意事项
- This recipe is LLM-orchestrated: read each phase, gather any missing inputs from the user, then call CLI commands. Use
muapifirst ifmuapi auth configureis unset.MUAPI_API_KEY - For model IDs without a CLI alias yet, fall back to the raw endpoint via and poll with
curl -X POST https://api.muapi.ai/api/v1/<endpoint> -H "x-api-key: $MUAPI_API_KEY" -H 'content-type: application/json' -d '{...}'.muapi predict wait <request_id> - Substitute placeholders with the user's actual inputs before issuing each call.
{{input_name}}
- 本流程由LLM编排:阅读每个阶段,向用户收集缺失的输入,然后调用CLI命令。如果
muapi未设置,先使用MUAPI_API_KEY进行配置。muapi auth configure - 对于尚未有CLI别名的模型ID,可通过原始端点回退,使用,并通过
curl -X POST https://api.muapi.ai/api/v1/<endpoint> -H "x-api-key: $MUAPI_API_KEY" -H 'content-type: application/json' -d '{...}'进行轮询。muapi predict wait <request_id> - 在发出每个调用前,将占位符替换为用户的实际输入。
{{input_name}}