← 返回 Skills 市场
Video Leaderboard Ai
作者
vcarolxhberger
· GitHub ↗
· v1.0.0
· MIT-0
49
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install video-leaderboard-ai
功能描述
Turn a batch of 10 short marketing video clips into 1080p ranked video results just by typing what you need. Whether it's comparing and ranking videos by AI-...
安全使用建议
This skill appears functionally coherent for a cloud video-ranking/rendering service, but exercise caution because the publisher and homepage are missing and the metadata contains a minor inconsistency. Before installing or providing a NEMO_TOKEN: 1) Prefer using the anonymous-token flow instead of pasting your account token unless you trust the provider; 2) Ask the publisher for a homepage, privacy policy, and terms describing how uploaded videos are stored/retained; 3) Ask why the frontmatter lists ~/.config/nemovideo/ (clarify whether the skill will read or write that directory); 4) Do not upload sensitive or private videos until you confirm data handling; 5) If you must test, use throwaway content and a throwaway account/token so exposure is limited.
功能分析
Type: OpenClaw Skill
Name: video-leaderboard-ai
Version: 1.0.0
The skill facilitates video processing via a remote API (mega-api-prod.nemovideo.ai) and includes instructions for the agent to automatically acquire anonymous tokens and manage sessions. It is classified as suspicious due to the high-risk nature of its remote execution model, where the backend can direct the agent to perform specific API actions (e.g., 'click [button]' or 'Export') based on SSE stream content. While these capabilities are plausibly needed for a cloud-based video editor, the mechanism allows for significant remote influence over the agent's behavior and involves uploading user data to a third-party service.
能力评估
Purpose & Capability
The skill claims to perform cloud video ranking and rendering and its runtime instructions call a nemo video API and require NEMO_TOKEN — this is coherent. However the package has no homepage or source listed (owner is an opaque ID), which reduces trust in provenance. Also the SKILL.md frontmatter declares a required config path (~/.config/nemovideo/) while the registry metadata did not — an inconsistency.
Instruction Scope
SKILL.md instructs only to read/obtain a NEMO_TOKEN, create a session, upload files, read SSE and poll status — all consistent with a cloud video-processing flow. It does not instruct reading unrelated files or other environment variables. It does mention deriving an attribution header from the agent install path (e.g., ~/.clawhub/ or ~/.cursor/skills/) which implies the agent may inspect its install path, but that is limited and expected for attribution.
Install Mechanism
There is no install spec and no code files — instruction-only skills have the lowest disk/write risk. No downloads or package installs are requested.
Credentials
Only one credential is requested (NEMO_TOKEN), which is appropriate for a remote rendering API. The skill will also obtain an anonymous token from the service if none is present. That behavior is plausible but important: providing an existing NEMO_TOKEN gives the skill access to your account on the service, so users should only supply it if they trust the provider. The SKILL.md frontmatter's configPaths (~/.config/nemovideo/) suggests possible filesystem access that is not otherwise referenced in the runtime instructions — this mismatch should be clarified.
Persistence & Privilege
The skill does not request always:true and defaults allow user invocation and autonomous invocation (the platform default). It does not request system-wide configuration changes or other skills' credentials.
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 video-leaderboard-ai - 安装完成后,直接呼叫该 Skill 的名称或使用
/video-leaderboard-ai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
v1.0.0
Initial release of Video Leaderboard AI — Rank and Compare Video Performance.
- Instantly rank and compare up to 10 short marketing video clips via AI performance metrics, with automated 1080p MP4 exports.
- Automatic setup: connects to a cloud backend, manages tokens, sessions, and lets users know when ready.
- Supports simple, natural prompts to upload, rank, export, and check credits without timeline editing or manual export settings.
- Accepts a wide range of video, audio, and image formats up to 500MB per file.
- Guides users through upload, ranking, credits, and export workflows; handles backend session and error logic transparently.
元数据
常见问题
Video Leaderboard Ai 是什么?
Turn a batch of 10 short marketing video clips into 1080p ranked video results just by typing what you need. Whether it's comparing and ranking videos by AI-... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 49 次。
如何安装 Video Leaderboard Ai?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install video-leaderboard-ai」即可一键安装,无需额外配置。
Video Leaderboard Ai 是免费的吗?
是的,Video Leaderboard Ai 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Video Leaderboard Ai 支持哪些平台?
Video Leaderboard Ai 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 Video Leaderboard Ai?
由 vcarolxhberger(@vcarolxhberger)开发并维护,当前版本 v1.0.0。
推荐 Skills