← Back to Skills Marketplace
489
Downloads
2
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install xhs-video-analyzer
Description
下载并分析小红书视频内容。当用户提供小红书链接（xiaohongshu.com）时，自动下载视频、提取语音文字、整理总结内容。Use when user provides a xiaohongshu.com URL and wants video content analysis.
Usage Guidance
This skill appears to implement the stated Xiaohongshu video → audio → Poe/Gemini transcription → summary flow, but double-check these before installing: (1) The published metadata omitted POE_API_KEY even though SKILL.md and the scripts require it — confirm you are comfortable the skill will read $POE_API_KEY or ~/.openclaw/openclaw.json. (2) The skill will upload audio to api.poe.com (a third‑party transcription service) — avoid using it with sensitive or private audio unless you trust the service and key. (3) It depends on ffmpeg (and optionally yt-dlp) being available; yt-dlp is used as a fallback. (4) The downloader scrapes the Xiaohongshu page and writes video/audio/transcripts into a work directory (default /tmp); check and clean that directory if needed. If you plan to use this skill, request the author to update the registry metadata to declare POE_API_KEY as a required environment variable and to document any other runtime requirements. If you don’t trust the source, consider inspecting or running the scripts in an isolated environment or using a throwaway/limited Poe API key.
Capability Analysis
Type: OpenClaw Skill
Name: xhs-video-analyzer
Version: 1.0.1
The skill provides a legitimate utility for downloading and analyzing Xiaohongshu videos by orchestrating ffmpeg, yt-dlp, and the Poe API. The scripts (full-analyze.sh, download.py) perform actions strictly aligned with the stated purpose in SKILL.md, such as extracting audio and sending it to a transcription service. While it accesses the local ~/.openclaw/openclaw.json file to retrieve the necessary POE_API_KEY, this behavior is explicitly documented and used only for the intended API communication with api.poe.com.
Capability Assessment
Purpose & Capability
Name/description match the code and instructions: scripts download a Xiaohongshu video, extract audio (ffmpeg/yt-dlp), and send audio to the Poe API (Gemini) for transcription and summarization. There are no unrelated cloud credentials or unexpected services requested.
Instruction Scope
SKILL.md and scripts describe and implement the expected workflow. The shell script and Python downloader operate only on the provided URL/work directory. The agent is instructed to send audio files to api.poe.com for transcription, which matches the stated design. The scripts read ~/.openclaw/openclaw.json as a fallback for POE_API_KEY.
Install Mechanism
No install spec is present (instruction-only with included scripts). No remote download/install of arbitrary code occurs; only local scripts are run. This is low-risk from an install mechanism standpoint.
Credentials
Metadata lists no required env vars, but both SKILL.md and scripts require a POE_API_KEY (or an entry in ~/.openclaw/openclaw.json). This mismatch is a meaningful omission: the skill will attempt to read the API key from your home config file or environment and will send audio data to Poe. The number and type of credentials requested (a single Poe API key) are proportionate to the task, but the omission from metadata is an incoherence that should be fixed/confirmed before use.
Persistence & Privilege
The skill does not request always:true and does not modify other skills or system-wide settings. It runs transiently and writes outputs to a working directory (default /tmp/xhs-analysis-<ts>), which is expected behavior.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install xhs-video-analyzer - After installation, invoke the skill by name or use
/xhs-video-analyzer - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
No changes detected in this version.
- Version 1.0.1 contains no modifications from the previous release.
v1.0.0
Initial release of xhs-video-analyzer.
- Downloads and analyzes Xiaohongshu videos via URL input.
- Extracts audio, transcribes speech using Poe API + Gemini.
- Provides a detailed structured summary of video content in Chinese.
- Supports segmenting audio for long videos (3-minute chunks).
- Outputs video, audio, transcript, and structured deep analysis.
Metadata
Frequently Asked Questions
What is Xiao Hong Shu Video Analyser?
下载并分析小红书视频内容。当用户提供小红书链接（xiaohongshu.com）时，自动下载视频、提取语音文字、整理总结内容。Use when user provides a xiaohongshu.com URL and wants video content analysis. It is an AI Agent Skill for Claude Code / OpenClaw, with 489 downloads so far.
How do I install Xiao Hong Shu Video Analyser?
Run "/install xhs-video-analyzer" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Xiao Hong Shu Video Analyser free?
Yes, Xiao Hong Shu Video Analyser is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Xiao Hong Shu Video Analyser support?
Xiao Hong Shu Video Analyser is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Xiao Hong Shu Video Analyser?
It is built and maintained by KINO (@hviktortsoi); the current version is v1.0.1.
More Skills