generateInsight() parsed against the strict aiInsightResponseSchema,
cross-checked that every recommendation's metricSource appeared in
citations[], and threw InsightSchemaError on a second failure. Its only
caller was runWithFallback(), which had no production caller — every
live surface runs runRawCompletionWithFallback. So none of it executed.
Wiring it into the live path was the alternative and it is not viable:
the live insightResultSchema has no citations[] field at all and treats
metricSource as optional, so the cross-check would have rejected
essentially every generation rather than tightening anything.
Removed: runWithFallback and its param/result types, generateInsight
and its private helpers, findUncitedRecommendations,
findRecommendationsMissingRationale, the never-thrown InsightSchemaError
and the isHardProviderFailure branch that excluded it, and
computeConfidence — a deterministic override that only ran inside the
dead wrapper.
Kept: buildRetryCorrectionMessage, which comprehensive-generate sends on
its own retry leg; the file it lived in is renamed to match what it now
holds. aiInsightResponseSchema stays as the shape that message
describes. Note the retry prompt asks for a stricter payload than the
live validator requires — that mismatch predates this and is left alone.
computeCitationCoverage is untouched and still live: it is a metric, not
a gate, and v1.30.24 wired it into the comprehensive post-parse step.
Chain-runner coverage the deleted tests carried (last-working cache,
empty-input guard) is ported onto runRawCompletionWithFallback, which
shares the machinery.
The coverage check hung off generateInsight, whose only caller runWithFallback
has no production caller left -- every live surface routes through
runRawCompletionWithFallback instead. The admin quality dashboard was reading a
metric nothing emitted. Widen the input to the structural shape both payloads
share and call it from the comprehensive post-parse step, so the annotation
reflects real generations.
Writing the locale tests found three gaps the English-only banks hid: Polish
verb suffixes fell outside the ASCII word class, a trailing word boundary after
an accented character never closed, and German subordinate clauses put the verb
last so the natural phrasing of a dose instruction never matched. The payload
collector also skipped legacy bare-string recommendations entirely.
After the schema parse and citation cross-check pass, the wrapper
counts how many recommendations make a normative claim ("target",
"should", "normal range", "above", "below") and how many of those
carry a `referenceId` pointing into the curated medical-reference
bundle. The result lands in the active Wide-Event meta annotation
under `ai_total_recommendations`,
`ai_normative_recommendations`,
`ai_cited_normative_recommendations` and
`ai_uncited_normative_recommendation_ids` so the admin AI quality
dashboard can chart citation-coverage over time.
Observational only in v1.4.16 — a normative rec without a referenceId
is logged, never raised as a parse error. Phase B5c will flip the
check to required for severity >= "important".
15 unit tests pin the keyword heuristic (EN + DE), the coverage
counts across cited / uncited / observational mixes, and the
`generateInsight()` integration that emits the annotation only on
success (skipped on schema-failure throw).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Marc-André Bombeck