media-ingest

Ingest video, audio, PDF, book, screenshot, and GitHub content into brain pages with backlinks.

Updated May 16, 2026

Search Tags:#ocr #backlinks #github #media #transcripts #ingest #entity-extraction

Ready to run

npx skills add https://github.com/Morris-UTRUST/GBrain-knowledge --skill media-ingest-morris-utrust

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: media-ingest
Source: https://github.com/Morris-UTRUST/GBrain-knowledge/tree/main/skills/media-ingest
Command: npx skills add https://github.com/Morris-UTRUST/GBrain-knowledge --skill media-ingest-morris-utrust

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ingests media content (video, audio, PDF, book, screenshot, and GitHub repo content) into the brain with analysis (not just a transcript) and automatic backlink propagation.

Core Features & Use Cases

Multi-format ingestion with automated transcription, OCR, and entity extraction.
Create brain pages by primary subject, with backlinks to people, companies, and sources.
Preserve raw sources for provenance and support deterministic filing workflows.

Quick Start

Ingest a media item by providing its URL or file path to automatically create a brain page with a concise summary and linked entities.

Frequently Asked Questions about media-ingest

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

FAQPage Schema

How do I ingest video and audio media into a searchable knowledge base?▼

To ingest video and audio media into a searchable knowledge base, the skill performs automated transcription, OCR, and entity extraction to create brain pages by primary subject with linked back-references. This process preserves raw sources for provenance and maintains a searchable history.

Can I extract entities and generate backlinks from PDFs and GitHub repositories?▼

Yes, you can extract entities and generate backlinks from PDFs and GitHub repositories. The skill ingests these formats to create brain pages by primary subject, automatically propagating backlinks to people, companies, and sources while preserving the original raw-source content.

What is the best way to index screenshots using OCR and entity extraction?▼

The best way to index screenshots using OCR and entity extraction is through multi-format ingestion workflows. The skill applies OCR to screenshot content, extracts entities, and generates a brain page with a concise summary and linked back-references for deterministic filing.

Does media ingestion support raw-source preservation and synchronized indexing?▼

Yes, media ingestion supports raw-source preservation and synchronized indexing. The workflow maintains provenance by preserving original files and ensures searchable history is kept synchronized across brain pages through automatic backlink propagation.

How do I create brain pages by primary subject from multi-format media?▼

You create brain pages by primary subject from multi-format media by providing a URL or file path. The skill automatically analyzes the content, generates a concise summary, extracts entities, and links back-references to people, companies, and sources.