wiki-ingest

Ingest local files, URLs, or pasted text into Herald's structured wiki memory.

Updated Mar 9, 2026
One-click install
npx skills add https://github.com/dbbaskette/herald --skill wiki-ingest-dbbaskette
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wiki-ingest
Source: https://github.com/dbbaskette/herald/tree/main/skills/wiki-ingest
Command: npx skills add https://github.com/dbbaskette/herald --skill wiki-ingest-dbbaskette

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ingests a local file, URL, or pasted text into Herald's long-term memory as a compounding wiki — creates a sources/ page with takeaways and creates or updates the related concepts/ and entities/ pages. Use whenever the user says ingest this, remember this article, add this doc to memory, save this URL, or hands over a file/link they want the knowledge from — even without the exact word ingest.

Core Features & Use Cases

  • Creates a structured memory entry (sources page) with takeaways, and cross-links to related concepts and entities for future recall.
  • Updates or creates related concepts and entities pages to reflect new information and maintain a connected knowledge graph.
  • Accepts inputs from local files, URLs, or pasted text and integrates them into Herald's long-term memory for reasoning.

Quick Start

Ingest this URL, file, or pasted text to create a new sources page and update related concepts and entities.

Frequently Asked Questions about wiki-ingest

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

FAQPage Schema
How do I save a URL or document into long-term memory for future recall?

To save a URL or document into long-term memory, ingest the source to automatically create a sources page with takeaways and update related concepts and entities pages for cross-linking and future recall.

What is the best way to build a connected knowledge graph from pasted text and local files?

Building a connected knowledge graph from pasted text and local files requires ingesting the inputs to generate structured wiki pages, cross-linking sources with related concepts and entities to enable reasoning across the data.

Can I add an article to memory and automatically generate takeaways and related concepts?

Yes, you can add an article to memory by ingesting the URL or text, which automatically generates a sources page containing takeaways and creates or updates related concepts and entities pages to maintain a connected wiki.

Does ingesting a source into memory update existing concepts and entities pages?

Ingesting a source into memory does update existing concepts and entities pages, reflecting the new information safely to maintain an accurate, cross-linked knowledge graph for future reasoning.

What inputs are supported for ingesting text into a structured wiki?

Ingesting text into a structured wiki supports arbitrary inputs including local files, URLs, and pasted text, integrating each into long-term memory to generate connected sources, concepts, and entities pages.