idea-ingest

Ingest links, articles, and tweets into a knowledge brain with analysis and cross-links.

Updated Aug 17, 2026
One-click install
npx skills add https://github.com/AMC-JTC/gbrain-1 --skill idea-ingest-amc-jtc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: idea-ingest
Source: https://github.com/AMC-JTC/gbrain-1/tree/main/plugin/skills/idea-ingest
Command: npx skills add https://github.com/AMC-JTC/gbrain-1 --skill idea-ingest-amc-jtc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Saving interesting links and ideas usually means losing them in a bookmarks folder with no context, no connection to what you already know, and no record of who wrote them. This Skill turns every shared link into an analyzed, cited, cross-linked page in your knowledge brain. ## Core Features & Use Cases - Structured Ingestion: Fetches articles, tweets, and documents, uploads the raw source for provenance, and files the page by primary subject (people, companies, concepts, or sources). - Mandatory Author Tracking: Creates or updates a people page for every author and builds bidirectional cross-links between sources, authors, and mentioned entities. - Brain-Connected Analysis: Produces analysis that connects new content to existing brain knowledge, flagging contradictions and relevant projects rather than just summarizing. - Use Case: You paste a link to a blog post about flywheel effects. The Skill fetches it, creates concepts/flywheel-effects, updates the author's people page, cross-links mentioned companies, and tells you how the post contradicts a note you saved last month. ## Quick Start Read this article and save it to my brain with analysis: paste the URL you want ingested.

Frequently Asked Questions about idea-ingest

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

FAQPage Schema
How do I save a link to my knowledge brain with analysis?

Share the URL and ask the agent to read or save it. The Skill fetches the content, uploads the raw source, creates a brain page filed by primary subject, and replies with analysis connecting it to existing brain knowledge.

How does the skill handle the author of an ingested article?

Author tracking is mandatory. The Skill searches the brain for an existing people page, creates one if missing, updates its timeline with the new publication, and cross-links the source and author in both directions.

What happens if the link is paywalled or the fetch fails?

The Skill saves a stub page containing the URL, metadata, and the failure reason, then tells you the content could not be fetched and asks you to paste the text manually.

Does ingesting a duplicate URL create a second page?

No. Before ingesting, the Skill searches the brain for the URL. If found, it merges new content into the existing page instead of overwriting it and tells you the item was already ingested.

Can I ingest a video or podcast link into the brain?

Only metadata can be ingested for video or podcast links unless a transcript is available. The Skill notes this limitation and asks you to provide a transcript for full ingestion.