When the whole-document parse + one retry still fail, report instructions now run
_resilient_report before the total-loss path. A brace/quote-aware scanner
(_extract_object_spans) recovers each recommendation object whether pretty-printed
across many lines or NDJSON one-per-line; a truncated tail gets a best-effort
_try_repair; _partition_items validates each recovered object against the T02 item
schema. Valid items survive (output_validated=True, partial=True), malformed/
over-maxItems items are quarantined with provenance (index, error, raw, reason),
capped at 20. Error locality now matches the unit of work: one bad item costs one
item, not the whole report.
Verified against the real 06-26 shape: 7 valid recommendations + a truncated tail
now recovers all 7 and quarantines the broken tail (previously the whole run was
discarded). Happy-path maxItems top-N enforcement is deferred to T04 (count caps).
Full suite: 215 passed, 1 skipped.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add safe action interpolation and for_each binding for rule fan-out, update the weekly SBOM definition, cover the new evaluation path, and reconcile activity-core scope/workplans for the State Hub sync.
The CUST-WP-0045 canary failed validation twice without leaving any
record of what the model actually returned. The warning logged only the
error message ($: missing required property 'summary'), not the JSON
shape that triggered it — so diagnosing required modifying code and
re-running. Log a 2KB preview of the offending raw output alongside the
error so the next failure of this shape is one grep away from diagnosis.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
