Semantic nearest-neighbor discovery
Find passages that discuss the same idea even when they do not share the same exact wording, vocabulary, or sentence structure.
Analytics
Public-facing analytics are intentionally limited for now. The vector index already supports richer research workflows that will be exposed more carefully in later iterations.
Cross-author thematic comparison
Compare how multiple Fathers approach prayer, grace, martyrdom, or the Eucharist without reading the full corpus line by line first.
Context-first quote expansion
Start from the highest-signal sentence, then move immediately into the surrounding paragraph so the text remains anchored in its original context.
Reusable retrieval for assistants
The same index can ground AI workflows with stable IDs, canonical links, and auditable passage expansion rather than opaque summaries.
Coming Soon
- Similarity maps showing which passages cluster together across authors and works.
- Topic-level analytics for recurring themes, sparse zones, and retrieval blind spots.
- Safer public dashboards for usage trends, latency, and coverage quality.
- Deeper visual exploration tools once they are hardened for open-internet traffic.
Why this is staged
Vector demos and raw inspection endpoints are useful during development, but they need tighter abuse controls, caching strategy, and presentation work before they should be treated as a public-facing product surface.