← 返回 Skills 市场
VPick AI Video Creator
作者
snoopyrain
· GitHub ↗
· v1.0.2
· MIT-0
108
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install vpick-video-creator
功能描述
All-in-one AI video production studio on a visual canvas. Generate videos (Kling 3.0, Veo 3.1, Sora 2, Runway, Grok, Midjourney Video), generate images (nano...
安全使用建议
This skill appears to do what it says (cloud-based end-to-end video creation) but it routes your prompts and uploaded media through vpick.10xboost.org and onward to many third-party providers. Before installing: (1) Verify the privacy/retention policy on the VPick site and confirm where your media and prompts are stored and for how long. (2) Treat your MCP link as a secret — understand where the connector stores it and who can access that token. (3) Clarify whether any providers (e.g., ElevenLabs STS) require you to supply separate API keys despite the doc's 'no separate API keys needed' claim. (4) Consider using limited-permission or throwaway account/credits for sensitive content. If the project requires strong confidentiality, avoid uploading sensitive media until you confirm data handling and credentials behavior.
功能分析
Type: OpenClaw Skill
Name: vpick-video-creator
Version: 1.0.2
The skill bundle provides instructions for an AI agent to interface with the VPick video production platform via the Model Context Protocol (MCP). It contains no executable code, only markdown documentation (SKILL.md) and metadata (_meta.json). The instructions are focused on legitimate video and audio generation workflows using third-party models like Kling, ElevenLabs, and Suno, and do not exhibit any signs of data exfiltration, malicious execution, or prompt injection. All operations are routed through the stated domain vpick.10xboost.org.
能力评估
Purpose & Capability
The name/description (AI video studio) align with the runtime instructions (create projects, upload images, run image/video generators, TTS, lip-sync, combine/export). It legitimately aggregates many third-party models via the VPick server (vpick.10xboost.org), which is consistent with the claimed capabilities.
Instruction Scope
The SKILL.md tells the agent to send prompts and uploaded media to the VPick server which then forwards them to multiple external providers. This is expected for a cloud-based video service, but the doc contains an internal inconsistency: it states 'no separate API keys or credentials are needed' (auth via MCP link) while one model row (ElevenLabs STS v2) notes 'Free (user API key)'. It's unclear whether some providers require you to supply separate keys or whether VPick proxies them — that ambiguity affects privacy and setup. The instructions otherwise do not ask the agent to read local files or environment variables beyond uploading user media and using the MCP link.
Install Mechanism
This is an instruction-only skill (no install spec, no code files). That minimizes on-disk installation risk; nothing is downloaded or executed by the skill itself.
Credentials
The skill declares no required env vars or credentials and relies on an MCP URL (embedded token) supplied by the user. That's proportionate for a connector pattern, but the SKILL.md's unclear statement about ElevenLabs STS and the phrase 'treat your MCP link like a password' mean the MCP link is effectively a secret that will be stored in the agent/connector — verify where/how that token is stored and who can access it. No other unrelated credentials are requested.
Persistence & Privilege
always is false and the skill is user-invocable. It does not request permanent presence or claim to modify other skills/configs. Normal autonomous invocation is allowed by default and not a unique concern here.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install vpick-video-creator - 安装完成后，直接呼叫该 Skill 的名称或使用
/vpick-video-creator触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.2
## vpick-video-creator 1.0.2 Changelog
- Added a new "Security & Data Handling" section to the documentation.
- Clarified authentication: all access is via MCP link with embedded token; no extra API keys needed.
- Explained data flow, storage, and billing (charges go to VPick credits; prompts/media are routed through VPick and sent to third-party providers).
- Emphasized no local credentials or environment variables are required.
- Updated prerequisite steps to reflect the streamlined security and connection process.
v1.0.0
VPick Video Creator v1.0.1
- Added core documentation (SKILL.md) covering supported video, image, and audio models, workflow, and usage instructions.
- Details prerequisites, MCP setup, canvas/project management, and end-to-end AI video production pipeline.
- Explains MultiShot and single-shot video generation, plus steps for voiceover, music, lip sync, and vocal separation.
- Lists model costs, limitations, and best-use scenarios for each tool integrated into the workflow.
- Includes clear code examples for nodes, connections, and generator commands.
元数据
常见问题
VPick AI Video Creator 是什么？
All-in-one AI video production studio on a visual canvas. Generate videos (Kling 3.0, Veo 3.1, Sora 2, Runway, Grok, Midjourney Video), generate images (nano... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 108 次。
如何安装 VPick AI Video Creator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install vpick-video-creator」即可一键安装，无需额外配置。
VPick AI Video Creator 是免费的吗？
是的，VPick AI Video Creator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
VPick AI Video Creator 支持哪些平台？
VPick AI Video Creator 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 VPick AI Video Creator？
由 snoopyrain（@snoopyrain）开发并维护，当前版本 v1.0.2。
推荐 Skills