← Back to Skills Marketplace
214
Downloads
1
Stars
0
Active Installs
11
Versions
Install in OpenClaw
/install ohyesai-mv
Description
Generate MV (Music Video) with AI-driven storyboarding, image generation, and video synthesis.
README (SKILL.md)
Skill: 音乐视频 (MV) 生成器 (MV Generator)
1. 技能描述
该技能允许用户上传音频并将其转化为具有特定视觉风格(如动漫、写实)和指定比例(横屏/竖屏)的音乐视频(MV)。
2. 预置条件与配置
- API Key 获取: 引导用户前往 https://ohyesai.com 获取。
- 环境变量: 必须配置
OHEYSAI_API_KEY。 - 文件限制: 音频文件大小不得超过 50MB。
3. 核心参数
在启动流程前,需确保获得以下参数:
audioFile: 用户上传的音频原始文件。style: 视频的视觉风格(如:动漫风格、赛博朋克、唯美古风等)。aspectRatio: 画面比例,仅限以下两个值:LANDSCAPE: 横屏 (16:9)PORTRAIT: 竖屏 (9:16)
4. 接口定义 (API Reference)
4.1 上传音频接口 (Upload)
- URL:
POST https://ohyesai.com/ohyesai-next/api/vio/skill/upload?apikey={{OHEYSAI_API_KEY}} - Payload:
multipart/form-data(字段名:file) - 功能: 将本地音频转换为
audioFileId。
4.2 提交任务接口 (Submit)
- URL:
POST https://ohyesai.com/ohyesai-next/api/vio/skill/mv-submit?apikey={{OHEYSAI_API_KEY}} - Request Body:
{ "audioFileId": "string", "style": "string", "aspectRatio": "LANDSCAPE | PORTRAIT" } - Response: 返回
taskId(String)。
4.3 查询任务接口 (Query)
- URL:
GET https://ohyesai.com/ohyesai-next/api/vio/skill/mv-query?apikey={{OHEYSAI_API_KEY}}&taskId={{taskId}} - Response Structure:
{ "code": 0, "data": { "state": "SUCCESS | RUNNING | FAIL", "data": "视频下载URL(仅SUCCESS时存在)" } }
5. 执行流程 (Workflow)
第一阶段:准备与校验
- 文件校验: 接收用户音频文件,检查大小。若超过 50MB,提示用户压缩后重新上传。
- 获取 FileId: 调用 上传音频接口。如果失败,告知用户服务器连接异常。
第二阶段:参数确认
- 如果用户未指定
style,主动询问:“您希望视频是什么风格?(例如:动漫、写实、3D、水墨等)”。 - 如果用户未指定
aspectRatio,询问:“您需要横屏(LANDSCAPE)还是竖屏(PORTRAIT)?”
第三阶段:提交与轮询
- 提交: 调用 提交任务接口,记录返回的
taskId。 - 轮询: 初始等待 10 秒,随后每隔 5-10 秒调用一次 查询任务接口。
- RUNNING: 继续等待,并向用户发送进度反馈(如:“正在为您精心制作视频,请稍候...”)。
- FAIL: 停止轮询,告知用户任务失败,并尝试分析原因。
- SUCCESS: 提取
data.data中的视频链接,并以预览卡片或链接形式发送给用户。
6. 异常与边界处理
- API Key 缺失: 若环境变量未配置,引导用户前往官网。
- 格式错误: 若上传非音频文件,拦截并提示。
- 超时处理: 若轮询超过 10 分钟仍为
RUNNING,提示用户记录taskId稍后手动查询。
7. 交互示例
User: 帮我把这首歌做成一个动漫风格的竖屏视频。 [附件: song.mp3]
Agent:
- (Internal) 检查 song.mp3 大小 (\x3C50MB)。
- (Internal) 调用
/upload获取audioFileId。- (Internal) 调用
/mv-submit发起任务,获得taskId: "f7fb7..."。- (Output) 已收到您的请求!正在为您生成动漫风格的竖屏 MV,这可能需要一点时间,我会在这里通知您进度。
- (Internal/Loop) 调用
/mv-query,状态为RUNNING。- (Internal/Loop) 调用
/mv-query,状态变为SUCCESS。- (Output) 视频制作完成!您可以点击下方链接查看或下载:[视频链接]
联系客服


Usage Guidance
This skill will upload any audio you provide to https://ohyesai.com using the OHEYSAI_API_KEY you supply. Before installing or using it:
- Confirm you trust ohyesai.com (review their privacy policy and terms) because your audio will be sent off-device.
- Avoid uploading sensitive or confidential audio.
- Be aware the API key is passed in the documented endpoints (appears as a query parameter), which can expose keys in logs or referrers — treat the key as sensitive and rotate it if compromised.
- The skill requires curl and sleep; no local code is installed.
- If you need stronger privacy, prefer on-device tools or a service with clear retention/deletion guarantees.
Capability Analysis
Type: OpenClaw Skill
Name: ohyesai-mv
Version: 0.1.0
The skill is a legitimate integration for the ohyesai.com music video generation service. It uses standard system utilities (curl, sleep) to upload audio files and poll for task completion via the service's API. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found in SKILL.md or _meta.json.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description (AI-driven MV generation) align with required resources: an API key for ohyesai.com and network-capable tools (curl) to upload audio and poll tasks. Required binaries (curl, sleep) and the single env var OHEYSAI_API_KEY are proportionate to the stated purpose.
Instruction Scope
SKILL.md instructs the agent to read the user's audio file, upload it as multipart/form-data to ohyesai.com, submit a task and poll for completion. This stays within the MV-generation purpose, but it does transmit user audio (and uses the API key) to a third-party service — users should be aware of that transfer and any privacy/terms implications.
Install Mechanism
No install spec and no code files (instruction-only) — minimal surface written to disk. This is low-risk and appropriate for a wrapper that delegates work to a web API.
Credentials
Only one required environment variable (OHEYSAI_API_KEY) is declared and used in the documented API calls. The requested secret is proportional to a hosted API service. No unrelated credentials or config paths are requested.
Persistence & Privilege
always is false and the skill has no install-time persistence or system-wide config modifications. It does not request elevated/system privileges or modify other skills.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ohyesai-mv - After installation, invoke the skill by name or use
/ohyesai-mv - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
- Updated all API endpoint URLs from https://dev.ohyesai.com to https://ohyesai.com.
- Clarified API key acquisition instructions to reference the main ohyesai.com site.
- No functional changes to workflow or parameters.
v0.0.8
No user-facing changes detected for version 0.0.8.
- Internal or documentation updates only.
- No changes in functionality, interface, or workflow.
v0.0.7
No changes detected in this version.
- Version 0.0.7 released with no modifications from the previous version.
v0.0.6
No user-facing changes in this release.
- No file changes detected.
- Functionality and documentation remain the same as the previous version.
v0.0.5
- Removed the _meta.json file from the skill package.
- No changes to the SKILL.md or core functionality.
v0.0.4
- Added a second customer service QR code image at the end of the documentation.
- No other changes or updates; all core skill logic and documentation remain the same.
v0.0.3
- Added _meta.json file for additional metadata.
- Appended a customer service contact section with QR code to the documentation.
v0.0.2
**Simplified workflow and interface for faster MV generation**
- Now requires only three core parameters: audio file, visual style, and aspect ratio (landscape/portrait).
- Workflow streamlined to three main steps: file upload, parameter confirmation, and submit/poll for results.
- MV creation process is now fully automated—no more multi-stage storyboarding, image or script confirmation.
- File size checks and user-friendly prompts added for error handling and missing information.
- New concise API definitions: one upload endpoint, one submit, and one query endpoint for easier integration.
- Enhanced user instructions and error messages for smoother experience.
v0.0.1
ohyesai-mv 0.0.1
- Initial release.
- Enables AI-driven music video (MV) creation, including storyboarding, image generation, and video synthesis via ohyesai.com API.
- Guides users through multi-stage, interactive MV production with asynchronous task handling and progress updates.
- Enforces strict file upload, polling, and user confirmation protocols.
- Delivers real video files to users after completion.
v1.0.1
ohyesai-mv 1.0.1
- Added a strict file upload requirement: All user-provided local files (audio, images) must be uploaded with the `uploadFile` API before further processing.
- Introduced the `uploadFile` tool and specified its usage and interface details.
- Updated SOP steps to require acquiring `fileId` via `uploadFile` before referencing files in all subsequent tools.
- Refined system prompt and workflow to enforce exclusive use of `fileId` (not file paths) in parameters like `audioFileId`, `imageFileId`, and `refImageFileId`.
- No code changes, only documentation improvements and stricter workflow guidance.
v1.0.0
ohyesai-mv 1.0.0 – Initial Release
- Launches a skill for generating full music videos (MV) using AI-driven storyboarding, image creation, and video synthesis.
- Supports multi-step, async MV workflow: reference image analysis, script generation, subject/environment image generation, video production, and audio/video merging.
- Enforces real-time progress updates and strict polling for async tasks; user is kept informed at each stage.
- Provides structured guidance and sample prompts for professional, director-style user interactions.
- Requires OHEYSAI_API_KEY for access and includes detailed setup and usage instructions.
Metadata
Frequently Asked Questions
What is music video mv 制作?
Generate MV (Music Video) with AI-driven storyboarding, image generation, and video synthesis. It is an AI Agent Skill for Claude Code / OpenClaw, with 214 downloads so far.
How do I install music video mv 制作?
Run "/install ohyesai-mv" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is music video mv 制作 free?
Yes, music video mv 制作 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does music video mv 制作 support?
music video mv 制作 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created music video mv 制作?
It is built and maintained by bajie-git (@bajie-git); the current version is v0.1.0.
More Skills