libingest

Name: libingest
Availability: InStock
Author: Copilot-LD

Convert PDFs, PowerPoints, and images into Schema.org annotated HTML.

1|1|Updated Aug 7, 2025

Search Tags:#automation #pdf #html #knowledge-management #ingest #schema.org #document-pipeline

Ready to run

npx skills add https://github.com/copilot-ld/copilot-ld --skill libingest

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: libingest
Source: https://github.com/copilot-ld/copilot-ld/tree/main/packages/libingest
Command: npx skills add https://github.com/copilot-ld/copilot-ld --skill libingest

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @copilot-ld/libllm, @copilot-ld/libpolicy, @copilot-ld/libprompt, @copilot-ld/libstorage, @copilot-ld/libtype, @copilot-ld/libutil, js-yaml.

What problem does it solve?

libingest provides a structured document ingestion workflow that converts PDFs, PowerPoints, and images into Schema.org annotated HTML for efficient indexing and knowledge extraction.

Core Features & Use Cases

Orchestrates a configurable pipeline (pdf-to-images, images-to-html, extract-context, annotate-html, normalize-html) to transform source documents into rich HTML.
Produces intermediate artifacts (context, image fragments, annotated HTML) for downstream retrieval and analysis.
Suitable for indexing knowledge bases, document search, and content catalogs.

Quick Start

Drop a document into the ingest folder and start the ingestion pipeline to generate structured HTML ready for indexing.

Frequently Asked Questions about libingest

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

FAQPage Schema

How do I convert PDFs to structured HTML for knowledge management indexing?▼

To convert PDFs to structured HTML for indexing, this Skill applies a configurable pipeline that transforms PDFs into images, then into HTML, and annotates them with Schema.org metadata for downstream retrieval.

Can I ingest PowerPoint presentations into Schema.org annotated HTML?▼

Yes, you can ingest PowerPoint presentations into Schema.org annotated HTML. The pipeline processes PowerPoints alongside PDFs and images, extracting context and normalizing the output for efficient document search.

What is the best way to automate document ingestion for a knowledge base?▼

The best way to automate document ingestion for a knowledge base is using a structured pipeline that handles pdf-to-images conversion, context extraction, and HTML annotation to produce searchable intermediate artifacts.

Does the document ingestion pipeline store intermediate artifacts for downstream retrieval?▼

Yes, the document ingestion pipeline stores intermediate artifacts including extracted context, image fragments, and annotated HTML, ensuring robust validation and enabling downstream retrieval and analysis.

What file formats are supported by the Schema.org HTML annotation pipeline?▼

The Schema.org HTML annotation pipeline supports PDFs, PowerPoint presentations, and images, converting them through steps like pdf-to-images, images-to-html, extract-context, annotate-html, and normalize-html.

Why use Schema.org annotated HTML for document indexing instead of raw text extraction?▼

Schema.org annotated HTML provides structured metadata that enhances document indexing and knowledge extraction, offering richer context and intermediate artifacts compared to basic raw text extraction from source files.