ingest-sources

Ingest and consolidate source documents into a unified memory with provenance.

283|44|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/kbanc85/claudia --skill ingest-sources
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ingest-sources
Source: https://github.com/kbanc85/claudia/tree/main/template-v2/.claude/skills/ingest-sources
Command: npx skills add https://github.com/kbanc85/claudia --skill ingest-sources

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ingests and unifies multiple source documents to prevent missing entities by inventorying sources, establishing provenance, and aligning memories across transcripts, emails, and documents.

Core Features & Use Cases

  • Inventory before processing to ensure complete coverage
  • File-Then-Extract workflow that preserves provenance
  • Consolidation and verification to deduplicate entities and ensure dedicated sources are represented

Quick Start

Ingest a folder of transcripts and emails to create a unified memory of key entities.

Frequently Asked Questions about ingest-sources

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

FAQPage Schema
How do I consolidate multiple transcripts and emails into a unified memory?

Consolidate multiple transcripts and emails by using a batch processing workflow that inventories sources, extracts entities, and deduplicates records to create a unified memory. This ensures complete coverage and aligns information across all documents.

What is provenance tracking and why do I need it for document ingestion?

Provenance tracking links extracted memories back to their originating documents during ingestion. You need it to ensure traceable memories and auditability, preventing missing entities and allowing you to verify the source of consolidated information.

Can I batch process a folder of mixed documents and emails to extract key entities?

Yes, you can batch process a folder of mixed documents, emails, and transcripts. The workflow applies inventory, extraction, consolidation, and verification phases to ensure all dedicated sources are represented and unified into a single memory.

What is the best way to deduplicate entities across multiple source documents?

The best way to deduplicate entities across multiple source documents is to apply a consolidation and verification phase after extraction. This process aligns memories and ensures dedicated sources are accurately represented without missing critical information.

Why are entities missing when I ingest multiple source documents without an inventory?

Entities are missing because without an inventory phase, there is no baseline to ensure complete coverage. Ingesting multiple source documents requires establishing provenance and inventorying sources first to prevent missing entities and align memories.