graph TD %%{init: {'theme': 'mc','layout': 'elk'}}%% APIRequest-pvhfg[Consulta feeds de un canal de Youtube] style APIRequest-pvhfg stroke:#a170ff ParseData-7xzz7[<div><img src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>Extraer el resultado XML] style ParseData-7xzz7 stroke:#a170ff JSONtoData-iw9ps[<div><img src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>JSON to Data] style JSONtoData-iw9ps stroke:#a170ff JSONtoData-qcvd4[<div><img src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>JSON to Data2] style JSONtoData-qcvd4 stroke:#a170ff ParseData-mmbzq[<div><img src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>extrae nuevas publicaciones] style ParseData-mmbzq stroke:#a170ff LanggraphDatabatchAgent-0gmtm[Data batch Agent] style LanggraphDatabatchAgent-0gmtm stroke:#a170ff youtube_transcripts-7xaub[YouTube Transcripts] style youtube_transcripts-7xaub stroke:#a170ff OpenAIModel-tzflz[<div><img src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>Análisis del xml] style OpenAIModel-tzflz stroke:#a170ff OpenAIModel-i2iir[<div><img src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>OpenAI2] style OpenAIModel-i2iir stroke:#a170ff TelegramMessage-elf5b[Telegram Message2] style TelegramMessage-elf5b stroke:#a170ff AdvancedAgent-pyhhs[Generador de publicidad] style AdvancedAgent-pyhhs stroke:#a170ff APIRequest-pvhfg -.- ParseData-7xzz7 linkStyle 0 stroke:#a170ff JSONtoData-iw9ps -.- ParseData-mmbzq linkStyle 1 stroke:#a170ff ParseData-mmbzq -.- JSONtoData-qcvd4 linkStyle 2 stroke:#a170ff JSONtoData-qcvd4 -.- LanggraphDatabatchAgent-0gmtm linkStyle 3 stroke:#a170ff ParseData-7xzz7 -.- OpenAIModel-tzflz linkStyle 4 stroke:#a170ff OpenAIModel-tzflz -.- JSONtoData-iw9ps linkStyle 5 stroke:#a170ff AdvancedAgent-pyhhs -.- LanggraphDatabatchAgent-0gmtm linkStyle 6 stroke:#a170ff youtube_transcripts-7xaub -.- AdvancedAgent-pyhhs linkStyle 7 stroke:#a170ff TelegramMessage-elf5b -.- AdvancedAgent-pyhhs linkStyle 8 stroke:#a170ff OpenAIModel-i2iir -.- AdvancedAgent-pyhhs linkStyle 9 stroke:#a170ff

New Video Promotion from a YouTube Channel

🧩 General Description

Starting from an RSS feed query of a YouTube channel, creates a personalized promotion and distributes the link of recent videos automatically. The system queries the channel's RSS feed, uses artificial intelligence to identify videos published in the last 24 hours, temporally classifies all publications, processes each new video in parallel to obtain its transcription translated to Spanish, generates inspiring summaries under 200 words, and distributes personalized promotions via Telegram, all in a fully automated workflow.

⚙️ Main Features

Automatic RSS feed query from YouTube channel
Intelligent AI analysis to detect new videos (last 24 hours)
Temporal classification: NEW (0-24h), RECENT (24-48h), OLD (>48h)
Obtaining video transcriptions translated to Spanish
Generation of inspiring summaries under 200 words with motivational tone
Automatic promotion distribution via Telegram
Parallel processing of multiple videos (up to 5 simultaneous)
Automatic detection of publication date and age calculation
Structured format for promotional messages

🔄 Workflow Steps

Component Name	Role	Inputs	Outputs
Feed Query	Obtains channel RSS	YouTube feed URL	XML with complete video list
Extract XML	Extracts feed content	API response	Raw XML text
XML Analysis	Analyzes and classifies videos	RSS structure XML feed	JSON with temporal video classification
Convert to Data	Transforms analysis JSON	Complete analysis JSON text	Structured data object
Extract New Publications	Filters only recent videos	Complete analysis data	Array with only publicaciones_nuevas
Convert to List	Prepares for batch processing	New publications JSON	List of individual video objects
Batch Agent	Distributes parallel processing	List of new videos	Coordinates up to 5 simultaneous videos
YouTube Transcriptions	Obtains video content	Video URL	Transcription translated to Spanish
Advertising Generator	Creates promotional summary	Transcription and video data	Inspiring summary under 200 words
Telegram Message	Distributes promotion	Summary, URL, date	Message sent to channel
OpenAI Analysis Model	Processes XML and detects new videos	RSS feed XML	JSON with classification and metadata
OpenAI Summary Model	Generates promotional content	System prompt and transcription	Summary text for Telegram

🧠 Notes

The system monitors the SET India channel (ID: UCAOgCnzXLTbaD-xSrVRDzOg) but can be configured for any channel
To obtain the channel ID: go to the channel page → View source code → search for "channel_id"
Analysis with OpenAI GPT-3.5-turbo uses temperature 0.1 for consistent responses
The analysis System Prompt is "# RSS FEED ANALYSIS - NEW PUBLICATION DETECTION"
Transcriptions are automatically translated to Spanish (translation_language: "es")
Summaries are designed to be inspiring, interesting and make the user feel special
Promotions are sent to the Telegram chat with ID 5632001849
Parallel processing allows handling up to 5 videos simultaneously (max_concurrency: 5)
Message format includes: [Incentive] + "Publication date: [date]" + "Watch it here: [URL]"
Analysis JSON contains: estado_analisis, total_publicaciones, nuevas_24h, publicaciones_nuevas[], ultima_publicacion
Each video in publicaciones_nuevas includes: video_id, titulo, publicado, hace_horas, url
Uses two AI models: one for RSS feed analysis (temporal detection) and another for summary generation (creativity)
RSS URL follows the format: https://www.youtube.com/feeds/videos.xml?channel_id={CHANNEL_ID}
Label components document: channel_id retrieval, detection flow, data transformations, batch agent architecture
The system is ideal for content creators who need to automatically promote each new publication