skills.rest
AMC-JTC
AMC-JTCCOMMUNITY
Explore all skills by this author →
GitHub

media-ingest

Ingest video, audio, PDF, book, screenshot, and GitHub repo content into a knowledge brain with entity extraction.

Updated Aug 17, 2026
Search Tags:#ocr#pdf#knowledge graph#youtube#transcription#entity extraction#media ingestion
One-click install
npx skills add https://github.com/AMC-JTC/gbrain-1 --skill media-ingest-amc-jtc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: media-ingest
Source: https://github.com/AMC-JTC/gbrain-1/tree/main/plugin/skills/media-ingest
Command: npx skills add https://github.com/AMC-JTC/gbrain-1 --skill media-ingest-amc-jtc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Valuable content locked in videos, podcasts, PDFs, books, screenshots, and GitHub repos stays disconnected from your knowledge base, forcing you to manually transcribe, summarize, and cross-reference every source. ## Core Features & Use Cases - Multi-Format Ingestion: Handles YouTube videos, audio files, PDFs, books, screenshots, and GitHub repositories with format-specific extraction (transcription, OCR, repo analysis). - Entity Extraction & Back-Linking: Automatically detects every person and company mentioned, creates or enriches their brain pages, and adds back-links and timeline entries. - Provenance Preservation: Uploads raw source files and files pages by primary subject rather than media format. - Use Case: Paste a YouTube link to a conference talk and get a brain page with a summary, timestamped highlights, and cross-linked pages for every speaker and company mentioned. ## Quick Start Ingest this YouTube video into my brain and link every person and company mentioned to their existing pages.

Frequently Asked Questions about media-ingest

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

FAQPage Schema
How do I ingest a YouTube video into a knowledge base?▼

Provide the YouTube URL as the source parameter. The skill fetches the transcript via captions or a transcription service, creates a brain page with a summary and timestamped highlights, and back-links every person and company mentioned.

How to summarize a book PDF into structured notes?▼

Pass the book PDF as the source. The skill extracts text, identifies chapters, and summarizes by chapter rather than inlining full text. For books over 500 pages, it links to the raw upload instead of embedding content.

Does media ingestion work with screenshots and images?▼

Yes, screenshots and images are processed through a vision model for OCR, extracting text and entities. Unclear segments are marked as illegible rather than guessed, and extracted entities are back-linked to brain pages.

What happens if transcription fails during audio ingestion?▼

If STT or captions are unavailable, the page notes that the transcript is unavailable and proceeds with whatever metadata exists. Content is never fabricated, and transient API failures are retried once before aborting.

How does the skill avoid creating duplicate pages?▼

Before creating a page, it searches the brain for the source URL or file hash. If a match exists, it asks whether to update the existing page or skip, preventing duplicates from re-running ingestion on the same source.

Related Skills

Explore More Personal & Entrepreneur Skills→
buildoakbuildoak
15

summarize

Extract and transcribe media content from URLs, local files, and streams.

Community
Advanced
weipingweiping
1

meeting-ingestion

Ingest meeting transcripts into structured knowledge pages with linked attendees and entities.

Community
Advanced
GYF0311GYF0311
5

meeting-ingestion

Convert meeting transcripts into structured pages with attendee and entity enrichment.

Community
Advanced
nbiishnbiish

w5-extractor

Extract W5 details from media and update Obsidian Knowledge Graph notes.

Community
Advanced