Accept traced entity capture citations and improve correction feedback

src/viano_bug_analysis/analysis/correction.py

@@ -23,7 +23,7 @@ MODEL_CODES = frozenset({"LLM_OUTPUT_SCHEMA", "LLM_RESPONSE", "LLM_INCOMPLETE_OU
     "INTERPRETATION_SCHEMA", "INTERPRETATION_SOURCE_UNKNOWN", "INTERPRETATION_SOURCE_KIND",
     "INTERPRETATION_EXPECTATION_UNKNOWN", "INTERPRETATION_CONTRADICTION_SOURCES",
     "INTERPRETATION_GAP_SOURCES", "INTERPRETATION_ENTITY", "INTERPRETATION_RESOLVED_ENTITY_OMITTED",
-    "INTERPRETATION_UNRESOLVED_ENTITY_OMITTED", "HYPOTHESIS_SCHEMA", "HYPOTHESIS_COUNT",
+    "INTERPRETATION_UNRESOLVED_ENTITY_OMITTED", "INTERPRETATION_ENTITY_PROVENANCE", "HYPOTHESIS_SCHEMA", "HYPOTHESIS_COUNT",
     "HYPOTHESIS_DUPLICATE", "HYPOTHESIS_SOURCE_UNKNOWN", "HYPOTHESIS_GROUNDING", "HYPOTHESIS_PRIMARY_FOCUS"})
 _attempt_context = contextvars.ContextVar("vba_early_correction", default=None)
 
@@ -158,6 +158,13 @@ def feedback(reader, phase, failure):
                         if len(available) <= 32:
                             issue["allowed_source_ids"] = available
                         issues.append(issue)
+        if error.get('prompt_version') == 'case_interpretation_resolved_v2':
+            from .contracts import entity_citation_issues
+            issues.extend(entity_citation_issues(payload,context))
+            for sid, source in context['sources'].items():
+                if (source['kind']=='data_quality' and source['case_pointer'].startswith('/entity_resolution/')
+                        and not any(sid in s.get('source_ids',[]) for s in payload.get('unknowns',[]) if isinstance(s,dict))):
+                    issues.append({'field':'unknowns','source_id':sid,'constraint':'retain_scoped_lookup_uncertainty'})
         issues = issues[:32]
     if phase == "hypotheses" and isinstance(payload, dict):
         from .hypothesis_contracts import _normalized
@@ -186,8 +193,10 @@ def feedback(reader, phase, failure):
                     if focus: break
                 if not set(proposal.get("interpretation_ids", [])) & focus:
                     issues.append({"field": "hypotheses/0/interpretation_ids", "constraint": "primary_focus", "allowed_source_ids": sorted(focus)[:32]})
-    return {"code": error["code"], "issues": issues[:32],
-            "instruction": "Return a complete replacement JSON object conforming to the supplied schema and source rules. Do not invent facts or sources. Use empty arrays/null for unsupported claims. Distinct hypotheses must retain valid grounding and count constraints."}
+    instruction = "Return a complete replacement JSON object conforming to the supplied schema and source rules. Do not invent facts or sources. Use empty arrays/null for unsupported claims. Distinct hypotheses must retain valid grounding and count constraints."
+    if phase == 'interpretation' and error.get('prompt_version') == 'case_interpretation_resolved_v2':
+        instruction += " Keep tester observations in observed_behavior with observation sources; move technical assertions to confirmed_technical_facts with fact sources. Retain selected entity_id values and cite their original mentions plus entity or matching capture evidence. Keep lookup uncertainties scoped to their queries."
+    return {"code":error["code"],"issues":issues[:32],"instruction":instruction}
 
 
 def augment(request, reader, phase, attempt):