en un clic
seedance-video
// 通过火山引擎 Ark API 生成 AI 视频 — 文生视频、图生视频、多模态、视频编辑、视频续写、长视频分镜拼接。任务异步执行+轮询,成功后可自动下载到本地。
// 通过火山引擎 Ark API 生成 AI 视频 — 文生视频、图生视频、多模态、视频编辑、视频续写、长视频分镜拼接。任务异步执行+轮询,成功后可自动下载到本地。
Bailian-powered unified creative studio — built-in Wan/Qwen image generation + HappyHorse 1.0 (text-to-video / image-to-video / reference-to-video / video-edit) + Wan 2.6/2.7 + 5 digital-human modes (photo speak / video relip / video reface / pose drive / avatar compose) + CosyVoice & Edge-TTS + storyboard long-video pipeline. Use when the user asks for AI image or video generation on Aliyun Bailian, keyframes, ecommerce images, multi-character video with audio sync, video editing/style transfer, talking-head, lip-sync, video reface, or storyboard-driven long video.
Web search and content extraction with Tavily and Exa via inference.sh CLI. Apps: Tavily Search, Tavily Extract, Exa Search, Exa Answer, Exa Extract. Capabilities: AI-powered search, content extraction, direct answers, research. Use for: research, RAG pipelines, fact-checking, content aggregation, agents. Triggers: web search, tavily, exa, search api, content extraction, research, internet search, ai search, search assistant, web scraping, rag, perplexity alternative
DashScope-powered digital human studio — photo speak, video relip, video reface, avatar compose. Use when the user asks for an AI talking-head video, lip-sync replacement on an existing video, replacing a person inside a video, or composing a new character from multiple reference images.
Guide AI-powered video editing, highlight extraction, silence removal, and talking-head polish through ClipSense.
Research, score, and plan content ideas through the Idea Research plugin.
Create, translate, repair, and burn subtitles through the Subtitle Craft plugin.
| name | seedance-video |
| description | 通过火山引擎 Ark API 生成 AI 视频 — 文生视频、图生视频、多模态、视频编辑、视频续写、长视频分镜拼接。任务异步执行+轮询,成功后可自动下载到本地。 |
| env_any | ["ARK_API_KEY"] |
输入一句中文/英文描述(可选参考图/视频),输出一段 AI 生成的高清视频。
所有调用都是 Ark API 异步任务:先 POST /contents/generations/tasks 拿 task_id,
再轮询 GET task_id,就绪后下载视频 URL 并落盘到 data_dir/videos/。
tongyi-image skill);纯文本/音频(用其他插件)seedance_create({prompt, model?, mode?, ratio?, duration?, resolution?, ...})seedance_edit({prompt, source_video_url, model?, duration?, ratio?, resolution?})seedance_extend({prompt, source_video_url, model?, duration?, ratio?, resolution?})seedance_transition({prompt, source_video_url, next_scene_prompt?, model?, duration?, ratio?, resolution?})seedance_status({task_id})seedance_list({limit?})| mode | 描述 | 典型模型 |
|---|---|---|
| t2v | 纯文本生视频 | seedance-1-lite |
| i2v | 首帧图生视频 | seedance-1-lite |
| i2v_end | 首尾帧图生视频 | seedance-1-lite |
| multimodal | 多图/视频/音频参考生成 | seedance-1-lite |
| edit | 视频编辑(替换/叠加) | seedance-1-lite |
| extend | 视频续写(向前/向后/链式) | seedance-1-lite |
prompt + (可选 image/video/audio) → Ark POST → task_id →
poll loop(默认 10s)→ status==completed →
extract video URL → (可选 auto_download) → 落盘 + 广播 task_update
故事脚本 → LLM 分镜拆解(decompose_storyboard) → N 段分镜 →
串行模式: 上一段末帧 → 下一段首帧参考 (视觉连贯)
云端续写模式: 上一段 video_url → Seedance extend 生成下一段/过渡段
并行模式: 各段独立生成 →
ffmpeg concat → 最终长视频(含转场)
edit 和 extend 必须使用 Seedance 任务返回的公网 video_url,通过 source_video_url 传入。/api/plugins/... 预览地址或 base64 视频当作源视频传给 edit / extend;Ark 无法访问这些本地资源。seedance_transition 表示“云端生成过渡/续写片段”,不是 ffmpeg 拼接。若需要最终长片,先生成过渡片段,再用本地 concat 合成。/long-video/concat 是本地合成:none 是硬拼接,crossfade 是本地 xfade 重编码,不等于 Seedance 云端 AI 过渡。| Gate | 检查内容 | 通过条件 |
|---|---|---|
| G1 | Ark API Key 已配置 | await tm.get_config("ark_api_key") 非空 |
| G2 | upload 路径在插件 data_dir 内(防遍历) | add_upload_preview_route 已注册 + path.relative_to(base) OK |
| G3 | task 异常用日志 + UI broadcast 兜底 | _broadcast_update(task_id, "failed") + error_message 落库 |
| 信任点 | 怎么自查 |
|---|---|
| 钱花在哪 / Cost | Ark API 唯一调用方;本插件不引入第三方计费 |
| 数据流向 / Data | 上传文件先存 data_dir/uploads/,仅作为 base64 发给 Ark API |
| 出错怎么办 / Errors | 全部异常落库 error_message,UI 通过 task_update 推送 |
| 远程依赖 / Network | 仅 https://ark.cn-beijing.volces.com/;无遥测 |
auto_download=true(默认开)才能离线复用video_url,不能传本地 mp4ArkClient 继承 BaseVendorClient,自动获得 retry/timeout/VendorError 分类on_unload 清理后台任务和资源,避免 Windows 文件锁问题api.spawn_task(...) 管理,unload 时统一 cancel + drain/storage/stats 走 SDK collect_storage_stats,UI 不再卡顿POST /upload 走 SDK add_upload_preview_route,防路径遍历