← 返回 Skills 市场
108
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install virtualover
功能描述
Turn raw video clips into polished, layered productions with virtualover — a skill built for blending virtual graphics, overlays, and composited elements dir...
安全使用建议
This skill appears to be a genuine NemoVideo frontend: it needs a Nemo token and will call NemoVideo endpoints and upload footage. Before installing, consider: (1) token leakage — the skill builds a claim URL that includes your token in the query string (this can leak via browser history or referrer headers); avoid using high-privilege or long-lived tokens if you proceed, or prefer the anonymous 7-day token flow and revoke it after use; (2) file writes and probing — the skill will create ~/.config/nemovideo/client_id and store session state, and it may check paths like ~/.clawhub and ~/.cursor/skills to detect platform; if you are uncomfortable with those filesystem accesses, do not install; (3) uploads — any video you give the skill will be transmitted to the NemoVideo backend, so do not upload sensitive or private footage unless you trust the service and have read its privacy policy. If you want to proceed, prefer using a short-lived/anonymous token and review NemoVideo’s docs for token revocation and data retention policies.
功能分析
Type: OpenClaw Skill
Name: virtualover
Version: 1.0.1
The virtualover skill is a legitimate integration for the NemoVideo API, designed to overlay graphics onto video footage. It follows standard OpenClaw patterns, including explicit metadata declarations for environment variables and configuration paths (~/.config/nemovideo/). The instructions in SKILL.md guide the agent through a standard authentication and session management flow with the official backend (mega-api-prod.nemovideo.ai) without any evidence of data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
Name/description, required env var (NEMO_TOKEN), and the runtime actions (uploading video, creating sessions, calling compositing endpoints) are coherent. Asking for a NemoVideo token is expected for a NemoVideo-backed overlay/rendering skill.
Instruction Scope
The SKILL.md instructs the agent to create and persist a ~/.config/nemovideo/client_id and to store session_id values; it also instructs detecting the install platform by checking paths like ~/.clawhub/ and ~/.cursor/skills/. These filesystem accesses go beyond the single declared configPath and are not fully declared. Critically, the skill instructs building a claim URL that embeds the token as a query parameter (https://nemovideo.com/workspace/claim?token=$TOKEN...), which risks leaking the token via browser history, referrers, or logs.
Install Mechanism
No install spec or downloaded code — instruction-only skill. This is the lowest install-risk category; nothing is written via an install step beyond what the instructions say at runtime.
Credentials
Only one credential (NEMO_TOKEN) is required, which is proportional to the stated purpose. The SKILL.md does offer to obtain an anonymous token by POSTing to the NemoVideo auth endpoint if NEMO_TOKEN is not set — that is reasonable but the instructions are ambiguous about how/where that token is stored and used, which affects privacy considerations.
Persistence & Privilege
The skill asks to write a client_id to ~/.config/nemovideo/ and to store session_id for subsequent calls. That local persistence is plausible for a client, but users should be aware state and tokens may be stored on disk. The skill is not marked always:true and does not request system-wide privilege changes.
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 virtualover - 安装完成后，直接呼叫该 Skill 的名称或使用
/virtualover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
v1.0.1
- Updated skill documentation for improved clarity and conciseness in onboarding and setup instructions.
- Refined descriptions and usage examples to better highlight key overlay features.
- Streamlined first-time connection steps and authentication explanation.
- Adjusted positioning and tone of introductory and guidance sections for accessibility.
- No changes to the code or skill logic; this is a docs-only update.
v1.0.0
Version 1.0.0 (Initial Release)
- Launches VirtualOver, enabling dynamic virtual overlays and animated elements on user videos with AI-driven compositing.
- Automatic setup: seamlessly manages NemoVideo API tokens, sessions, and user authentication.
- Accepts standard video formats (mp4, mov, avi, webm, mkv) and supports image/audio uploads.
- Provides intuitive commands for overlays, BGM, credits, status, export, and more, with intelligent request routing.
- Robust error handling and session management ensures reliability for all users.
- Designed for creators, marketers, and educators seeking high-quality, layered video without complex editing tools.
元数据
常见问题
Virtualover 是什么？
Turn raw video clips into polished, layered productions with virtualover — a skill built for blending virtual graphics, overlays, and composited elements dir... 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 Virtualover？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install virtualover」即可一键安装，无需额外配置。
Virtualover 是免费的吗？
是的，Virtualover 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Virtualover 支持哪些平台？
Virtualover 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 Virtualover？
由 tk8544-b（@tk8544-b）开发并维护，当前版本 v1.0.1。
推荐 Skills