corpus-familiarization

Automates initial familiarization of new RAG reference material into knowledge artifacts.

Updated Dec 4, 2025

Search Tags:#rag #pipeline #artifact-generation #knowledge-map #reference-material #familiarization

Ready to run

npx skills add https://github.com/jamieadams-nerd/umrs-project --skill corpus-familiarization

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: corpus-familiarization
Source: https://github.com/jamieadams-nerd/umrs-project/tree/main/.claude/skills/corpus-familiarization
Command: npx skills add https://github.com/jamieadams-nerd/umrs-project --skill corpus-familiarization

SYSTEM DOCUMENTATION & REQUIREMENTS

Corpus Familiarization Skill

Builds active, always-on knowledge from RAG source material. Produces four artifacts that function as the agent's internal map of what it knows — so it can reason about the corpus before issuing any retrieval query.

Frequently Asked Questions about corpus-familiarization

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

FAQPage Schema

How do I build a knowledge map from new RAG reference material?▼

To build a knowledge map from RAG reference material, automate the initial familiarization process to generate structural artifacts. This creates an active, always-on internal map so the agent can reason about the ingested corpus before issuing any retrieval queries.

What is corpus familiarization in a RAG pipeline?▼

Corpus familiarization in a RAG pipeline is the process of mapping newly ingested reference material into a persistent knowledge store. It produces four artifacts that function as an internal inventory, ensuring the agent knows exactly what source documents it has available.

When should I run corpus familiarization after a document ingestion?▼

You should run corpus familiarization immediately after ingesting new documents across collections and after any significant corpus updates. This ensures the agent's active knowledge map accurately reflects the latest reference material before querying.

How do I maintain a running inventory of ingested reference material?▼

You maintain a running inventory of ingested reference material by writing four structural artifacts to a persistent knowledge store during corpus familiarization. This inventory allows the agent to reference what it knows without scanning the raw documents.

Does corpus familiarization work without external dependencies?▼

Yes, corpus familiarization operates without external dependencies, relying solely on your existing RAG pipeline. It processes newly ingested documents and writes four artifacts directly to your persistent knowledge store to maintain an active mapping.

Why does my RAG agent fail to reason about newly ingested collections?▼

A RAG agent fails to reason about newly ingested collections because it lacks an active knowledge map of the corpus. Running an automated familiarization process writes reference artifacts to a persistent store, giving the agent the structural context needed for accurate retrieval.