video-analyzer

Convert video URLs or local paths into Chinese analysis notes with [mm:ss] timestamps.

15|4|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/Ceeon/video-analyzer-skill --skill video-analyzer-ceeon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: video-analyzer
Source: https://github.com/Ceeon/video-analyzer-skill/tree/main
Command: npx skills add https://github.com/Ceeon/video-analyzer-skill --skill video-analyzer-ceeon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the effort of watching a video manually by generating a detailed, time-stamped understanding of what is said and shown throughout the timeline.

Core Features & Use Cases

  • Video download & compression: Fetches videos from supported platforms (e.g., Bilibili/YouTube) or uses a local file, then compresses to an analysis-friendly size (e.g., 720p, CRF 28).
  • Files API upload & Responses API analysis: Uploads the processed video to the Files API to obtain a file_id, then runs the Responses API using an input_video reference.
  • Structured notes with [mm:ss] timestamps: Produces structured notes in Chinese with accurate time markers, key points, on-screen text, formulas, scenes, and actions.

Quick Start

Ask Claude Code to analyze this video link or local path: “分析这个视频 https://www.bilibili.com/video/BVxxx 并输出带[mm:ss]时间戳的详细笔记(中文)。”

Frequently Asked Questions about video-analyzer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate timestamped notes from a Bilibili or YouTube video?

Yes, you can analyze local video files by providing the file path instead of a URL. The analyzer processes the local file through ffmpeg compression and the Files API to generate timestamped Chinese analysis notes.

How does the video analysis pipeline handle large video files?

The analyzer uses the Responses API with an input_video reference and file_id to parse the video deterministically. This mechanism ensures accurate extraction of key points, on-screen text, formulas, scenes, and actions into structured notes.

Do I need yt-dlp and ffmpeg installed to analyze videos?

The analyzer is best suited for long-form content where understanding detailed explanations, summarizing key scenes, and extracting structured takeaways with timeline timestamps are needed. It outputs comprehensive Chinese notes rather than brief overviews.

What limitations exist when using the Files API for video analysis?

A key limitation of using the Files API for video analysis is that videos must be compressed to an analysis-friendly size beforehand. Extremely long videos might lose fine details during the 720p compression step before processing.