skills.rest
atlasgroupcz
atlasgroupczOFFICIAL
Explore all skills by this author →
GitHub

video-analyze

Transcribe and analyze video and audio files or YouTube URLs using Gemini multimodal models.

2|1|Updated Mar 30, 2026
Search Tags:#multimodal#youtube#gemini#timestamps#speech to text#audio analysis#video transcription
One-click install
npx skills add https://github.com/atlasgroupcz/codexis-marketplace --skill video-analyze-atlasgroupcz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: video-analyze
Source: https://github.com/atlasgroupcz/codexis-marketplace/tree/main/plugins/video-analyze/skills/video-analyze
Command: npx skills add https://github.com/atlasgroupcz/codexis-marketplace --skill video-analyze-atlasgroupcz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Getting usable text or answers out of video and audio content normally requires juggling tools like yt-dlp, ffmpeg, or whisper, which may not be available in the environment. This Skill provides a single command that transcribes media files and YouTube URLs with Gemini multimodal models and answers questions about their content. ## Core Features & Use Cases - Timestamped Transcription: Generate detailed transcripts with spoken dialogue, scene descriptions, and audio cues from local files (mp4, webm, mov, avi, mkv, mp3, wav, flac, ogg, aac) or YouTube URLs. - Direct Question Answering: Query the media directly with a specific question when the transcript lacks a detail, such as a visual element at a given timestamp. - Use Case: You have a recording of a team meeting in mp4 format. Generate a timestamped transcript, read it to summarize the key decisions, and locate the exact minute where pricing was discussed. ## Quick Start Ask the assistant to transcribe your video or audio file, for example: transcribe the meeting recording at /home/codexis/meeting.mp4 with timestamps.

Frequently Asked Questions about video-analyze

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

FAQPage Schema
How do I transcribe a video file to text with timestamps?▼

Run the video-analyze transcript command with the file path as the source. It prints the path of the saved transcript file, which contains timestamped dialogue, scene descriptions, and audio cues produced by Gemini Pro.

Can I transcribe a YouTube video without downloading it?▼

Yes, pass the full YouTube URL (youtube.com or youtu.be) directly as the source to the transcript command. The media is sent to Gemini for transcription without requiring yt-dlp or youtube-dl.

What audio and video formats are supported for transcription?▼

Supported video formats are mp4, webm, mov, avi, and mkv. Supported audio formats are mp3, wav, flac, ogg, and aac. YouTube URLs are also accepted as input sources.

Why does video transcription time out or fail?▼

The default 30-second shell timeout is too short because Gemini upload and multi-pass transcription regularly take 30 seconds to 3 minutes. Always pass timeoutMs of 300000 to the shell call so the transcript file is written to disk.

What should I do when I get a 429 rate limit error?▼

Rate limit errors are transient. Wait 10 to 30 seconds and retry the same command; the transcription should complete on a subsequent attempt.

How do I ask about a visual detail not in the transcript?▼

Use the direct query mode by passing the source and a specific question, such as asking about the color of a logo at a given timestamp. This re-sends the media to Gemini with your targeted question.

Related Skills

Explore More Content & Communication Skills→
Clawdi-AIClawdi-AI
4

openai-whisper

Transcribe local audio files to text or SRT using the Whisper CLI.

Official
Basic
jonmagicjonmagic
4

archive-meeting

Convert Zoom and Teams transcripts into linked Transcript, Executive Summary, and Meeting Notes artifacts.

Community
Advanced
garochee33garochee33
3

transcribe

Transcribe audio and video recordings into text with speaker diarization.

Community
Intermediate
mangiapanejohn-devmangiapanejohn-dev
34

openai-whisper

Transcribe audio files to text locally using the Whisper CLI.

Community
Basic