Bible reference lookups, full-text/topic/Strong's/Q&A search, and AI semantic search in
Raamattu Nyt. Use for fetching verses by reference (Joh 3:16), building/modifying the
SearchPage (7 tabs: jakeet/aiheet/Strong's/kysymykset/videot/muut/comparison), calling search
RPCs (search_text, search_topics_with_aliases, search_verses_by_strongs,
get_approved_ai_summary), classifying queries (verse vs topic vs Strong's vs free-text),
anchor questions/AI summaries/topic synonyms, validating OSIS references, and choosing the
right lookup pattern.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Um comando direto ignora o prompt de revisão. Verifique a origem antes de executá-lo.
Bible reference lookups, full-text/topic/Strong's/Q&A search, and AI semantic search in
Raamattu Nyt. Use for fetching verses by reference (Joh 3:16), building/modifying the
SearchPage (7 tabs: jakeet/aiheet/Strong's/kysymykset/videot/muut/comparison), calling search
RPCs (search_text, search_topics_with_aliases, search_verses_by_strongs,
get_approved_ai_summary), classifying queries (verse vs topic vs Strong's vs free-text),
anchor questions/AI summaries/topic synonyms, validating OSIS references, and choosing the
right lookup pattern.
Bible Lookup Helper
Covers two related but distinct workflows:
Lookup — fetch a specific verse / chapter / range when the reference is already known
Search — surface results across verses, topics, Strong's, questions, and AI summaries when the user types a free-form query
When in doubt which one applies: a known reference (Joh 3:16) → lookup; a free-form query (rakkaus, armo ja totuus) → search. The search path itself classifies the query and picks the right backend.
Context Files (Read First)
For schema and API details, read from Docs/context/:
Before debugging search RPC performance or behavior, skim references/learnings.md (terminology conventions + pointers to cross-cutting DB-perf entries in .claude/LEARNINGS.md).
Detailed Search Reference
For the full search architecture — 7 tabs, per-modality RPCs, query classification, AI-summary caching/feedback, anchor questions, recent additions — read references/search-system.md.
Read it when the task touches:
The SearchPage UI or any *SearchSection.tsx component
A search RPC that's not in the basic lookup list below
AI semantic search, summary caching, or thumbs-up/down feedback
Generate correct RPC function calls for Bible lookups
Validate OSIS reference formats
Suggest optimal lookup patterns (verse vs chapter vs range vs search)
Provide book name aliases and normalization
Convert between different reference formats
Route free-form queries to the right search RPC (verses / topics / Strong's / Q&A / AI)
Reason about the AI-semantic-search flow end-to-end: cache key (normalizeQueryKey), ai_summary_feedback table, ai-semantic vs ai-searchfeature_key, quality circuit-breaker (thumbs_down ratio), and the verse-text rendering chain (AISemanticResults → fetchVerseTexts → mapFinnishToOsis → get_verses_by_ref)
When to delegate to other skills
This skill owns the SEARCH side of AI semantic search (cache, routing, rendering, feature_key selection at the call site). For deeper AI configuration, hand off to ai-prompt-manager:
Question / task
Skill
"Why is the AI summary card blank for <book> <ch>:<v>?"
