‹ 首页

muapi-one-shot-video

@samuraigpt · 收录于 1 周前 · 上游提交 3 周前

Generate a single continuous cinematic shot video — no cuts, one seamless flowing scene with dramatic lighting and motion.

适合你,如果需要快速生成无剪辑、有电影感的单镜头视频

/ 下载安装
muapi-one-shot-video.skill双击,或拖进 Claude 桌面版 / Cowork,即完成安装↓ .skill↓ .zip
用别的 agent?下载 .zip 解压,把文件夹放进它的技能目录
Claude Code~/.claude/skills/(项目级 .claude/skills/)
Codex CLI~/.codex/skills/
Cursor自动读取上面两处目录
其他工具见其文档的「skills」目录;两个下载是同一份文件,只是名字不同
/ 通过 npx 安装 校验哈希
npx oh-my-skill add samuraigpt/generative-media-skills/muapi-one-shot-video
/ 通过 bash 安装
curl -fsSL https://oh-my-skill.com/install.sh | bash -s -- samuraigpt/generative-media-skills/muapi-one-shot-video
/ 已经装过?验证本机副本,不用重装
npx oh-my-skill verify samuraigpt/generative-media-skills/muapi-one-shot-video
安装目标可用 --agent / --scope 或 --to 明确指定;省略时只会在唯一已存在的 agent 目录上自动选择,零命中或多命中会停止并提示。content_hash 缺失或不一致均拒装。
3754GitHub stars
~586上下文体积 · 单文件
镜像托管

怎么用

商店整理自技能原文 · 版本 a1c4c98 · 表述以原文为准
它做什么

安装后,Claude 能根据你描述的场景、风格、时长和画面比例,生成一段连续不剪辑的电影感视频。如果提供参考图,还能让视频内容与图片保持一致。

什么时候触发

当你提到“一镜到底”、“单次拍摄”、“连续视频”等关键词,或要求生成一段无剪辑的连续镜头视频时触发。

装好后可以这样说
Claude 会调用视频生成工具,输出一段连续镜头。
Claude 会基于参考图生成指定时长和比例的连续镜头。
技能原文 SKILL.md作者撰写 · MIT · a1c4c98

One-Shot Video

Generate a single continuous cinematic shot video — no cuts, one seamless flowing scene with dramatic lighting and motion.

Inputs

| Name | Type | Required | Default | Description | |:---|:---|:---|:---|:---| | scene | text | yes | — | The scene to render (e.g. "a chef plating food in a moody Michelin-star kitchen"). | | style | text | no | cinematic, anamorphic lens, shallow depth of field, dramatic lighting | Visual tone and style (e.g. "noir, handheld, golden hour"). | | duration | text | no | 10s | Target duration (e.g. "5s", "10s"). | | aspect_ratio | text | no | 16:9 | Output aspect ratio — "16:9", "9:16", or "1:1". | | reference_image | image_url | no | — | Optional reference image for subject/scene anchoring. |

Steps
Phase A — Generate the One-Shot Video

Submit the plan with ONE step:

  1. One-shot video — If {{reference_image}} is provided, use muapi video generate (model=veo3.1-image-to-video); otherwise use muapi video generate (model=veo3.1-text-to-video).
  2. Prompt: {{scene}}, one continuous uncut shot, no transitions, camera slowly moves through scene, {{style}}, ultra cinematic, film grain, 4K quality
  3. Aspect ratio: {{aspect_ratio}}
  4. Duration: {{duration}}

After generation, present the video and suggest:

  • A second angle variation
  • Adding ambient sound with mmaudio-v2-video-to-video
Notes
  • Emphasize "no cuts, no transitions" in the prompt for true one-shot feel.
  • For portrait/vertical style (9:16), add "vertical format, smartphone framing" to the prompt.
  • If the scene involves a person, suggest kling-v3.0-pro-image-to-video as an alternative for better human motion.
Trigger Keywords

one shot video, single take, continuous video, one take, cinematic shot, seamless video


Notes for the Executing Agent
  • This recipe is LLM-orchestrated: read each phase, gather any missing inputs from the user, then call muapi CLI commands. Use muapi auth configure first if MUAPI_API_KEY is unset.
  • For model IDs without a CLI alias yet, fall back to the raw endpoint via curl -X POST https://api.muapi.ai/api/v1/<endpoint> -H "x-api-key: $MUAPI_API_KEY" -H 'content-type: application/json' -d '{...}' and poll with muapi predict wait <request_id>.
  • Substitute {{input_name}} placeholders with the user's actual inputs before issuing each call.
按 MIT 许可原样转载,未经改动 · 在 GitHub 查看 →

评论

登录即可评论;带「已验证安装」的,是发布者名下有本店的安装或持有记录。