Vidilearn — Extrator de Conteúdo do YouTube
Alfo-Tech-Lab/vidilearn
Extraia transcrições e metadados do YouTube para seus AI agents — sem API keys
Ver no GitHubMCP server para extração de transcripts, legendas, capítulos e metadados estruturados de vídeos do YouTube sem necessidade de API keys. Ideal para alimentar pipelines RAG, agents de IA e sistemas de automação com dados estruturados em JSON.
Instalação
npx convez-ai Alfo-Tech-Lab/vidilearn1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Alimentar pipelines RAG com conteúdo de vídeos
- Treinar agents de IA com conhecimento extraído de YouTube
- Automação de coleta de dados para análise e contexto
- Integração com MCP servers e sistemas de automação
- Geração de datasets estruturados para aprendizado de modelos
Veredito

Ferramenta prática e leve para extração de conhecimento do YouTube sem overhead de APIs. Bem integrada ao ecossistema MCP e útil para quem precisa alimentar agents de IA com conteúdo estruturado de vídeos.
README
A tradução para o português deste README ainda está na fila.
Vidilearn
Teach your AI using YouTube videos
Lightweight AI-first YouTube extraction CLI for transcripts, subtitles, chapters, descriptions, and structured metadata — without API keys.
Installation • Quick Start • Features • AI Workflows • CLI Usage • JSON Output
Overview
Vidilearn is a modern developer-first CLI designed for AI agents, RAG pipelines, automation systems, MCP servers, Codex CLI workflows, Gemini CLI integrations, and educational tooling.
Extract structured knowledge from YouTube videos directly into your AI systems.
No API keys.
No browser automation.
No bloated setup.
Features
- Extract transcripts
- Download subtitles
- Parse descriptions
- Extract chapters & timestamps
- Structured JSON output
- AI-ready data formatting
- Fast CLI workflow
- No API key required
- Lightweight architecture
- Automation-friendly
- Works with AI agents & MCP systems
- Clean terminal experience
Installation
Global Installation
npm install -g vidilearn
Verify Installation
vidilearn --help
Quick Start
Extract Video Knowledge
vidilearn extract "https://youtube.com/watch?v=VIDEO_ID"
Rule of Thumb
Whenever a URL contains special characters like ?, &, or =, always wrap it in quotes.
Incorrect
vidilearn extract https://youtube.com/watch?v=abc123&list=xyz
Correct
vidilearn extract "https://youtube.com/watch?v=abc123&list=xyz"
Example JSON Output
{
"title": "Build AI Agents",
"channel": "AI Academy",
"duration": "12:45",
"description": "Learn how to build AI agents...",
"transcript": "...",
"chapters": [
{
"title": "Introduction",
"timestamp": "00:00"
},
{
"title": "Agent Architecture",
"timestamp": "03:42"
}
]
}
CLI Usage
Basic Extraction
vidilearn extract "<youtube-url>"
Save Output
vidilearn extract "<youtube-url>" > output.json
Pretty JSON Output
vidilearn extract "<youtube-url>" --pretty
Transcript Only
vidilearn extract "<youtube-url>" --transcript
Chapters Only
vidilearn extract "<youtube-url>" --chapters
Metadata Only
vidilearn extract "<youtube-url>" --metadata
AI & Agent Workflows
Vidilearn is designed for modern AI ecosystems.
Compatible With
- OpenAI Agents
- MCP Servers
- Codex CLI
- Gemini CLI
- Claude Workflows
- LangChain
- LangGraph
- CrewAI
- AutoGen
- RAG Pipelines
- Vector Databases
- Local AI Systems
- AI Automation Agents
MCP Server Integration
Use Vidilearn inside MCP-based AI systems to feed YouTube knowledge directly into your agents.
Example
vidilearn extract "<youtube-url>" --json
Pipe the structured output into:
- memory systems
- vector stores
- retrieval pipelines
- autonomous agents
Codex CLI Workflow
Use Vidilearn to give coding agents contextual learning data from technical YouTube videos.
Example
vidilearn extract "https://youtube.com/watch?v=VIDEO_ID" > context.json
Then feed the extracted data into:
- Codex CLI
- AI coding agents
- autonomous developer workflows
Gemini CLI Integration
Use Vidilearn as a knowledge ingestion layer for Gemini-powered automation systems.
Example Workflow
vidilearn extract "<youtube-url>" --pretty
Pass the extracted transcript into:
- Gemini CLI prompts
- summarization pipelines
- AI research systems
- educational assistants
AI Use Cases
RAG Pipelines
Convert long-form educational videos into searchable knowledge bases.
AI Memory Systems
Store YouTube knowledge into persistent AI memory.
Educational Applications
Turn lectures into structured AI-readable datasets.
Autonomous Agents
Allow AI agents to learn directly from YouTube.
Research Systems
Extract technical insights from conferences, tutorials, and talks.
Architecture Philosophy
Vidilearn is built around:
- simplicity
- speed
- AI-first workflows
- lightweight tooling
- structured extraction
- automation compatibility
Why Vidilearn?
Most YouTube extraction tools:
- require API keys
- are bloated
- break frequently
- are not AI-focused
- have poor automation support
Vidilearn focuses on:
- developer experience
- AI-native workflows
- structured outputs
- clean CLI ergonomics
- production-ready automation
Performance Goals
- Fast extraction
- Minimal dependencies
- Low memory footprint
- Automation-safe architecture
- Reliable structured output
Roadmap
- Streaming transcript extraction
- Batch playlist processing
- MCP-native server mode
- Local embedding pipeline support
- Vector database integrations
- AI summarization modules
- Multi-language subtitle support
- Live stream support
Contributing
Contributions are welcome.
Ideas, improvements, bug reports, and feature requests are appreciated.
Security
Vidilearn does not require API keys or authentication tokens.
Always review extracted content before using it in production AI systems.
License
MIT License
Support
If Vidilearn helps your workflow, consider sponsoring development ❤️
GitHub Sponsors: https://github.com/sponsors/sarathi-eng
Author
Built by Alfo Tech Industries
© 2026 Alfo Tech Industries. All rights reserved.