aitoearn-dsh-plugin-test 0.0.1

package/skills/video-generation/SKILL.md

@@ -0,0 +1,61 @@
+---
+name: video-generation
+description: Generate, edit, or extend videos with Seedance models via list_video_generation_models and generate_video; optimizes prompts per model's conventions. Use when user asks to create, edit, extend, or storyboard a video, or mentions seedance / text-to-video / image-to-video / video-to-video.
+requiredTools:
+  - list_video_generation_models
+  - generate_video
+---
+
+# Video Generation
+
+Generate, edit, or extend videos with `list_video_generation_models` and `generate_video`.
+
+The catalog is the source of truth for every video model. For Seedance families, use `references/seedance-2-0-guide.md` (default model is Seedance 2.0 Fast); use `references/seedance-2-5-guide.md` only when routed to 2.5. For other families, use only catalog capabilities and the user's intent; do not transfer Seedance-only prompt conventions.
+
+## Tools
+
+- `list_video_generation_models`: call first — before routing. Select the returned `id`; inspect `family`, `modes`, `input.frames`, `input.references`, `output`, and `billing`. The metadata fields are flat; there is no nested `metadata` object.
+- `generate_video` (v2): `model` + `prompt` required; `mode` is `generate`, `edit`, or `extend`; optional `frames: {first, last}`, `references: {images, videos, audios}`, `output: {resolution, aspectRatio, duration, seed, watermark}`, `options`, and `groupId`.
+  - `frames.first` / `frames.last` are first/last frame anchors (not ordinary references); `references.images` are ordinary reference images. Do not mix anchors with references.
+  - Async tool: `submit` creates the task, `resolveResult` polls `status` / `videoUrl` / `coverUrl`. Wait for completion before reporting.
+  - Credit estimate runs automatically; if the user must confirm, wait. Do not retry a rejected cost.
+
+Do not call other video APIs, CLIs, or local scripts.
+
+## When to use
+
+- Text-to-video, image-to-video, multi-reference video generation
+- Video editing (replace / add / remove elements or sound)
+- Video extension (forward / backward continuation)
+- First-frame / first-last-frame controlled generation, keyframe / storyboard generation
+
+## When not to use
+
+- Static image generation — use `image-creation` instead
+- SVG / vector icons, diagrams, or deterministic code-native output
+- Polling task status outside the `generate_video` async lifecycle
+
+## Decision tree
+
+1. **Intent**: generation / editing / extension / first-last-frame? Editing keywords: `编辑/修改/替换/增加/删除`; extension keywords: `延长/延续/续写/向前/向后`.
+2. **Model routing**: choose from the complete catalog. Default to Seedance 2.0 Fast (fastest and cheapest; multi-image ≤9 / multi-video ≤3 / multi-audio ≤3 references, first/last frames, basic edit & extend). Need 1080p / 4K → Seedance 2.0 standard. Duration > 15s, references beyond 2.0 limits (图 >9 / 视频 >3 / 音频 >3), timestamps, precise edit / seamless extend / keyframes / storyboard / whitebox / multi-view → 2.5. See `references/model-comparison.md`.
+3. **Input state**: classify per the Seedance guide (2.0 Fast by default, 2.5 when escalated) — 纯文生 / 需要参考但未提供素材 / 素材可读取 / 无法读取 / 规格预检. Hard limits (30 图, 10 视频/30s, 10 音频/30s, 合计 50) are 2.5 ceilings; for 2.0 Fast use the model's own `input` limits from the catalog (图 9 / 视频 3 / 音频 3, 视频音频总 15s). Probe readability and real resolution/duration with `get_media_metadata` before classifying provided material — do not guess.
+4. **Roles**: label every input media per the guide's `建立素材职责和主体映射` — image: first-frame / last-frame / reference; video: edit master / extend base / reference; audio: BGM / dialogue / SFX.
+
+## Workflow
+
+1. Call `list_video_generation_models` and pick `model` from `id`, plus a supported `mode` and `output` (`resolution`/`aspectRatio`/`duration`). Frame capability is indicated by `input.frames`; ordinary media limits live under `input.references`.
+2. Collect prompt, constraints / avoid list, and input media URLs. Label every image / video / audio role; enforce hard limits from `input` constraints; note stability guidance (1–8 img subjects, 1–5 av subjects 5–10 s). Use `get_media_metadata` to probe provided material (readability, resolution, duration) against the hard limits before submitting.
+3. Run the 8-step workflow per `seedance-2-0-guide.md` (default; `镜头1/2/3` instead of timestamps) or per `seedance-2-5-guide.md` when routed to 2.5 (timestamp control, free aspect ratio, multi-view): ① parse goal (story contract + dialogue ledger + required-entity list) → ② compile long text → ③ two-pass material understanding → ④ duty mapping + `【未采用素材】` → ⑤ confidence routing → ⑥ sole task + aspect-ratio gate (`请注意这不是视频编辑。` only on gate trigger) → ⑦ param rules → ⑧ template purification.
+4. Select template family: base / reference / multi-material / long-video / editing (sole master + range closure + motion-slot) / extension (boundary anchoring + single continuous instance) / keyframe (`@图片N作为首帧。` verbatim) / storyboard (≤15) / whitebox / emotion-camera-audio.
+5. Compose the prompt per template — replace all `<占位符>`. Do not write `output.aspectRatio` / `output.duration` / `output.resolution` into the prompt. Do not auto-add quality/watermark/subtitle constraints unless user requested.
+6. Call `generate_video` once per requested video. Pass anchors in `frames: {first, last}` and ordinary references in `references: {images, videos, audios}`.
+7. Wait for the async task to complete, verify `videoUrl` / `coverUrl` and check subject, style, and motion quality.
+8. Iterate with one targeted change if needed. Re-apply invariants on every retry.
+9. Report the returned `videoUrl`, final prompt, model, and params. Include `coverUrl` / `mediaId` when present.
+
+## References
+
+- `references/seedance-2-0-guide.md` — default methodology guide: 11 principles, 5 input states, 8-step workflow, all templates, output contract, 26-point self-check; `镜头1/2/3` instead of timestamps, fixed 6 ratios, single-view, text generation patches (1080p/4K is 2.0 standard-only; Fast is 480p/720p)
+- `references/seedance-2-5-guide.md` — same methodology; use when routed to 2.5 (timestamps, >15s, larger reference sets, precise editing, keyframes/storyboard/whitebox, multi-view)
+- `references/model-comparison.md` — 2.0 Fast (default) / 2.0 standard / 2.5 capability table and routing decision tree

package/skills/video-generation/references/model-comparison.md

@@ -0,0 +1,76 @@
+# Seedance 模型对比与选型
+
+Seedance 2.0 vs 2.5 差异速查及 Agent 选型路由。
+
+---
+
+## 能力对比
+
+| 维度 | Seedance 2.0（含 Fast） | Seedance 2.5 |
+|------|-------------|-------------|
+| 时间控制 | `镜头1/2/3` 分镜，不响应 `0-3s` 时间戳 | 响应 1s 时间戳 `0-3s` / `第5s` / 相对时间 |
+| 多视图 | 不建议，单视图更稳定 | 支持多视图 |
+| 宽高比 | 固定 6 档 | 固定 6 档（16:9 / 4:3 / 1:1 / 3:4 / 9:16 / 21:9） |
+| 素材上限 | 图 9 / 视频 3 / 音频 3（视频音频单段 2–15s，总 15s） | 图 30 / 视频 10 / 音频 10（视频音频总 30s） |
+| 分辨率 | Fast：480p / 720p；标准版：至 1080p / 4K | 480p / 720p |
+| 单段时长 | 4–15s | 4–30s |
+| 价格 | Fast 最便宜（720p ≈8.1 积分/秒）；标准版 720p ≈13.5、4K ≈68.6 积分/秒 | 720p ≈20.5 积分/秒 |
+| 文字生成 | 支持广告语 / 字幕 / 气泡台词 | 未重点支持 |
+| 编辑能力 | 基础（增删改） | 专业级（精准局部修改、风格转换） |
+| 延长能力 | 基础（向前/向后） | 专业级（无缝衔接） |
+| 关键帧/分镜 | 基础分镜 | 关键帧锚点 / 宫格分镜（≤15 格） / 白模参考 |
+| 生成速度 | Fast 最快 | 相对较慢 |
+
+## 选型路由
+
+### 默认 Seedance 2.0 Fast 的场景
+
+- **默认模型**：速度最快、价格最低，支持多图（≤9）/ 多视频（≤3）/ 多音频（≤3）参考、首尾帧、基础编辑与延长，满足大多数需求
+- 快速原型验证
+- 需要文字生成（广告语、字幕、气泡台词）
+
+### 升级到 Seedance 2.0 标准版的场景
+
+- 需要 1080p / 4K 分辨率（Fast 仅 480p / 720p，其余输入能力与标准版相同）
+
+### 需要升级到 Seedance 2.5 的场景
+
+- 视频时长 > 15s（2.0 系单段 4–15s，2.5 最长 30s）
+- 参考素材超出 2.0 系上限：图 > 9、视频 > 3、音频 > 3，或视频/音频总时长 > 15s
+- 需要时间戳精确控制（`0-3s` 区间、时间点转场）
+- 需要精准编辑（局部修改、风格转换、元素增删）或无缝延长（声画一致性要求高）
+- 需要关键帧锚点 / 宫格分镜 / 白模参考 / 多视图（2.0 系有限或不建议）
+
+### 决策流程
+
+```
+1. 用户是否指定模型？ → 是 → 直接使用该模型
+2. 需要 1080p / 4K？ → 是 → Seedance 2.0 标准版
+3. 时长 > 15s 或素材超出 2.0 系上限（图 >9 / 视频 >3 / 音频 >3）？ → 是 → 2.5
+4. 需要时间戳 / 精准编辑 / 无缝延长 / 关键帧 / 宫格分镜 / 白模 / 多视图？ → 是 → 2.5
+5. 其余情况 → Seedance 2.0 Fast（默认）
+```
+
+> **注意**：以上为 prompt 写法层面的选型建议。最终可用模型、分辨率、时长档位以 `list_video_generation_models` 返回的 catalog 为准，不要硬编码模型名。
+
+---
+
+## Prompt 写法差异速查
+
+| 写法 | 2.0 | 2.5 |
+|------|-----|-----|
+| 分镜标识 | `镜头1：...` `镜头2：...`（禁止 `0-3s`） | `0-3s：...` `3-7s：...` 或 `镜头1` |
+| 时间精度 | 不支持时间戳，时长由接口控制 | 支持 1s 精度时间戳（仅用户提供时使用） |
+| 多图指代 | 单视图推荐：`参考图片1中的主体` | 多视图支持：`图片1-2是人物1` |
+| 约束词 | `保持无字幕` / `不要生成Logo` | 同左（2.5 支持更细粒度音频控制） |
+| 特殊字符 | `()` 音乐 `<>` 音效 `{}` 台词 `【】` 字幕 | 同左 |
+| 首尾帧 | `图片1作为首帧` | `图片1作为首帧` + 首帧锁定比例 |
+| 方法论 | 11 原则 / 8 步流程 / 模板 / 输出契约 — 两模型完全一致 | 同左，仅时间/比例/多视图/分辨率/文字生成有差异 |
+
+## 校验清单（选型后）
+
+- [ ] 已调用 `list_video_generation_models` 确认目标模型存在
+- [ ] `output.aspectRatio` / `output.duration` / `output.resolution` 仅作为接口参数，不写入 prompt
+- [ ] 素材数量符合目标模型的 `input.frames`、`input.references` 及 `maxTotalFiles`
+- [ ] 时间控制写法与目标模型匹配（2.0 用镜头序号，2.5 可用时间戳）
+- [ ] 多视图仅在 2.5 中使用