Exercises: Build a Source Registry
Use starter.py to classify source documents and decide whether they are ready for ingestion.
1 Classify source format
Identify markdown, PDF, HTML, DOCX, CSV, JSON, API, or database sources.
2 Add governance fields
Track owner, authority, version, access level, freshness rule, and retirement status.
3 Decide ingestion readiness
Approve, quarantine, reject, or require review.
4 Explain the production risk
Return the first risk a RAG engineer should care about.