← 返回 Skills 市场
Video Finder
作者
peandrover adam
· GitHub ↗
· v1.0.0
· MIT-0
69
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install video-finder
功能描述
find search query into matched video results with this skill. Works with MP4, MOV, AVI, WebM files up to 500MB. content creators use it for searching a video...
安全使用建议
This skill is mostly coherent with its stated purpose (it uploads videos and uses a cloud rendering/search API), but you should be cautious: installing/using it will send your video files and metadata to the nemovideo backend (https://mega-api-prod.nemovideo.ai). Verify the service's privacy and terms before sending sensitive content. Also note a metadata mismatch: the SKILL.md references a config path (~/.config/nemovideo/) and detecting an install path for attribution headers — ask the publisher why the skill needs filesystem/install-path access and where session tokens are stored. If you don't want uploads or external processing of private videos, do not enable this skill.
功能分析
Type: OpenClaw Skill
Name: video-finder
Version: 1.0.0
The video-finder skill is a legitimate tool designed to interface with a cloud-based video processing API (mega-api-prod.nemovideo.ai). It provides the AI agent with clear instructions for managing authentication via the NEMO_TOKEN, handling file uploads, and processing video search queries. The skill includes security-conscious instructions, such as explicitly forbidding the agent from printing raw tokens or JSON to the user, and its environment-sniffing for attribution is limited to its own installation path. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found.
能力评估
Purpose & Capability
The name/description match the runtime instructions: the SKILL.md explains uploading videos, running searches, streaming SSE, and requesting cloud renders via https://mega-api-prod.nemovideo.ai. Requesting a single API credential (NEMO_TOKEN) is proportionate to that purpose. However, the frontmatter metadata includes a required config path (~/.config/nemovideo/) while the registry metadata earlier listed no config paths — this mismatch is unexplained.
Instruction Scope
The instructions direct the agent to: obtain or generate a token (POST to the external auth endpoint), open a session, upload video files, stream SSE, poll render state, and return download URLs. These behaviors are consistent with a cloud-rendering video search service, but they also mean user video files and metadata will be transmitted to an external third party. The SKILL.md also instructs the agent to detect an install path to set X-Skill-Platform (which implies reading filesystem/install paths). The doc says not to print tokens/raw JSON, but it does not explicitly instruct how long to persist session tokens or where to store them.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest install risk. There is no downloading or unpacking of third-party code indicated.
Credentials
Only one credential is declared (NEMO_TOKEN), which fits the API usage. The SKILL.md also offers generating an anonymous token via the service's auth endpoint if NEMO_TOKEN is not set (this is reasonable for anonymous use). The unexplained metadata configPath (~/.config/nemovideo/) and the requirement to detect install path for attribution headers are proportionally suspicious because they suggest possible filesystem access beyond just reading an env var; the SKILL.md does not justify reading local config files.
Persistence & Privilege
The skill does not request always:true and does not include an install script. It asks the agent to save a session_id during the session (expected). There is no instruction to modify other skills or system-wide settings.
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-finder - 安装完成后，直接呼叫该 Skill 的名称或使用
/video-finder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
Video Finder — Initial Release
- Launches the Video Finder skill for fast AI-powered video search and export.
- Supports searching a video library by topic or description; processes and returns 1080p MP4 videos in seconds.
- Works with MP4, MOV, AVI, WebM files up to 500MB; various file types accepted for upload.
- Handles user intents such as upload, export, balance check, and timeline state queries.
- Integrates with a cloud backend supporting anonymous token creation, session management, credits, state polling, and export workflows.
- Includes robust error handling and tips for best performance.
元数据
常见问题
Video Finder 是什么？
find search query into matched video results with this skill. Works with MP4, MOV, AVI, WebM files up to 500MB. content creators use it for searching a video... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 69 次。
如何安装 Video Finder？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install video-finder」即可一键安装，无需额外配置。
Video Finder 是免费的吗？
是的，Video Finder 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Video Finder 支持哪些平台？
Video Finder 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 Finder？
由 peandrover adam（@peand-rover）开发并维护，当前版本 v1.0.0。
推荐 Skills