ingest

Route incoming content to ingestion workflows by detecting type and extracting entities.

1|Updated May 9, 2026
One-click install
npx skills add https://github.com/weiping/gbrain-cn --skill ingest-weiping
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ingest
Source: https://github.com/weiping/gbrain-cn/tree/main/skills/ingest
Command: npx skills add https://github.com/weiping/gbrain-cn --skill ingest-weiping

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the manual effort of deciding what to save and how to structure it by automatically parsing incoming content and routing it into the right brain pages with citations and cross-links.

Core Features & Use Cases

  • Content Router & Delegation: Detects the input type (meetings, articles, media, documents, conversations) and applies the correct ingestion workflow.
  • Entity Detection with Cross-Referencing: Extracts people, companies, concepts, and events; creates/updates pages and enforces back-links (“Iron Law”) so mentions never go unlinked.
  • Timeline + Provenance Preservation: Writes reverse-chronological timeline entries with mandatory inline [Source: ...] citations and preserves raw sources for verifiability using gbrain files upload-raw.

Quick Start

Ask your agent to ingest a meeting transcript you provide so it can create the meeting page, update all referenced entity pages, and add timeline entries with proper source citations.

Frequently Asked Questions about ingest

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

FAQPage Schema
How do I automatically extract entities and citations from meeting transcripts?

To extract entities from meeting transcripts, route the content through an ingestion workflow that detects input types and extracts people, companies, and events. The system automatically creates pages, enforces back-links, and writes timeline entries with mandatory inline source citations.

What is timeline propagation when building a knowledge graph from documents?

Timeline propagation in a knowledge graph is the process of writing reverse-chronological entries across all entities mentioned in your documents. It ensures every event is tracked with inline source citations and raw source preservation for verifiability.

Can I ingest articles and PDFs into a knowledge graph with automatic back-linking?

Yes, you can ingest articles, web pages, and PDFs into a knowledge graph with automatic back-linking. The workflow detects the document input type, extracts concepts and entities, and enforces back-links so mentions never go unlinked across pages.

Does content ingestion require inline source citations for every written fact?

Yes, content ingestion requires mandatory inline source citations for every written fact added to the knowledge graph. The workflow enforces provenance preservation and saves raw sources via file upload to ensure verifiability of extracted entities and events.

What is the best way to route social posts and media transcripts into a knowledge brain?

The best way to route social posts and media transcripts into a knowledge brain is using a content router that detects the input type and delegates to the correct ingestion workflow. This extracts entities, applies timeline propagation, and preserves raw sources.