@claritylabs/cl-sdk 3.2.0 → 3.2.2

package/dist/index.mjs

@@ -5168,6 +5168,74 @@ function nodesByParent(sourceTree) {
   }
   return byParent;
 }
+function sourceNodeIdsBySpanId(sourceTree) {
+  const bySpan = /* @__PURE__ */ new Map();
+  for (const node of sourceTree) {
+    for (const spanId of node.sourceSpanIds) {
+      const nodes = bySpan.get(spanId) ?? [];
+      nodes.push(node.id);
+      bySpan.set(spanId, nodes);
+    }
+  }
+  return bySpan;
+}
+function operationalEvidenceScore(span) {
+  const text = cleanText([
+    span.text,
+    span.formNumber,
+    span.sourceUnit,
+    span.metadata?.elementType,
+    span.metadata?.sourceUnit
+  ].filter(Boolean).join(" "), "");
+  if (!text) return 0;
+  let score = 0;
+  if (span.sourceUnit === "table_row" || span.sourceUnit === "table") score += 5;
+  if (span.metadata?.elementType === "title") score += 4;
+  if (span.sourceUnit === "page") score -= 3;
+  if (/\b(policy\s*(number|period|term)|effective date|expiration date|expiry date|named insured|insurer|carrier|security|broker|producer|premium|total due)\b/i.test(text)) score += 12;
+  if (/\b(coverage part|limit(?:s)? of liability|deductible|retention|retroactive date|aggregate|sublimit|sub-limit|each claim|each loss|each occurrence|coinsurance)\b/i.test(text)) score += 14;
+  if (/\bendorsement\s+(?:no\.?|number|#)?\s*[A-Z0-9]|forms? and endorsements?|attached at inception|schedule\b/i.test(text)) score += 8;
+  if (/\bitem\s+\d+\.?\s*(?:named insured|policy number|policy period|renewal|form of business|coverage parts?|premium|extended reporting|producer|forms? and endorsements?)\b/i.test(text)) score += 10;
+  if (/\$[\d,.]+|[0-9]{1,2}\/[0-9]{1,2}\/[0-9]{2,4}|[0-9]{1,2}\s+[A-Za-z]{3,9}\s+[0-9]{4}/.test(text)) score += 3;
+  return score;
+}
+function operationalProfileEvidence(sourceTree, sourceSpans) {
+  const sorted = [...sourceSpans].sort(
+    (left, right) => (spanPageStart(left) ?? Number.MAX_SAFE_INTEGER) - (spanPageStart(right) ?? Number.MAX_SAFE_INTEGER) || (left.location?.charStart ?? Number.MAX_SAFE_INTEGER) - (right.location?.charStart ?? Number.MAX_SAFE_INTEGER) || left.id.localeCompare(right.id)
+  );
+  const selected = /* @__PURE__ */ new Set();
+  for (let index = 0; index < sorted.length; index += 1) {
+    const score = operationalEvidenceScore(sorted[index]);
+    if (score < 8) continue;
+    const page = spanPageStart(sorted[index]);
+    for (let offset = -2; offset <= 2; offset += 1) {
+      const neighborIndex = index + offset;
+      const neighbor = sorted[neighborIndex];
+      if (!neighbor || spanPageStart(neighbor) !== page) continue;
+      const neighborText = cleanText(neighbor.text, "");
+      if (!neighborText || neighborText.length > 5e3) continue;
+      selected.add(neighborIndex);
+    }
+  }
+  const nodeIdsBySpanId = sourceNodeIdsBySpanId(sourceTree);
+  const seenText = /* @__PURE__ */ new Set();
+  return [...selected].sort((left, right) => left - right).map((index) => sorted[index]).filter((span) => span.sourceUnit !== "table_cell").flatMap((span) => {
+    const text = cleanText(span.text, "");
+    if (!text) return [];
+    const key = `${spanPageStart(span) ?? "na"}:${text.toLowerCase().slice(0, 240)}`;
+    if (seenText.has(key)) return [];
+    seenText.add(key);
+    return [{
+      sourceSpanId: span.id,
+      sourceNodeIds: [...new Set(nodeIdsBySpanId.get(span.id) ?? [])].slice(0, 4),
+      pageStart: spanPageStart(span),
+      pageEnd: spanPageEnd(span),
+      sourceUnit: spanSourceUnit(span),
+      formNumber: span.formNumber,
+      text: text.slice(0, span.sourceUnit === "page" ? 1200 : 900)
+    }];
+  }).slice(0, 180);
+}
 function sourceTreeRootId(sourceTree) {
   return sourceTree.find((node) => node.kind === "document")?.id;
 }
@@ -5192,8 +5260,9 @@ function emptyOperationalProfile() {
     warnings: []
   };
 }
-function buildOperationalProfilePrompt(sourceTree) {
-  const nodes = operationalProfilePromptNodes(sourceTree).map(
+function buildOperationalProfilePrompt(sourceTree, sourceSpans) {
+  const evidence = operationalProfileEvidence(sourceTree, sourceSpans);
+  const fallbackNodes = evidence.length ? [] : operationalProfilePromptNodes(sourceTree).map(
     (node) => compactNode2(node, node.kind === "page" || node.kind === "endorsement" ? 900 : 700)
   );
   return `Extract a source-backed operational profile for an insurance policy or quote.
@@ -5204,7 +5273,8 @@ Return only high-value operational facts needed for policy lists, Q&A, complianc
 - coverage type labels
 
 Rules:
-- Every returned value must include sourceNodeIds or sourceSpanIds from the provided nodes.
+- Every returned value must include sourceNodeIds or sourceSpanIds from the provided evidence.
+- When citing an evidence entry, copy its sourceSpanId into the returned sourceSpanIds array.
 - If a value is not directly supported, omit it.
 - Prefer declarations, schedules, premium tables, and endorsement schedules over generic policy wording.
 - For broker/producer, extract the agency or company legal name, not the license role, credential, or type. In a block like "Bayshore Insurance Brokers, LLC" followed by "Surplus Lines Broker - CA License No. ...", broker.value must be "Bayshore Insurance Brokers, LLC"; the surplus-lines role and license number are not the broker name.
@@ -5220,10 +5290,10 @@ Rules:
 - For coverage schedules, put each claim, aggregate, sublimit, retention, deductible, and retroactive date values in coverages[].limits with labels and source IDs. Keep the legacy coverages[].limit as the primary display value only.
 - Extract coinsurance, participation percentage, or insurer/named-insured split terms as coverages[].limits entries with kind "other" when they are part of a coverage schedule.
 - Do not copy entire policy wording into fields.
-- Extract facts directly from source nodes. There is no deterministic fact baseline.
+- Extract facts directly from source evidence. There is no deterministic fact baseline.
 
-Source nodes:
-${JSON.stringify(nodes, null, 2)}
+Source evidence:
+${JSON.stringify(evidence.length ? evidence : fallbackNodes, null, 2)}
 
 Return JSON for the operational profile.`;
 }
@@ -5483,7 +5553,7 @@ async function runSourceTreeExtraction(params) {
     const response = await safeGenerateObject(
       params.generateObject,
       {
-        prompt: buildOperationalProfilePrompt(sourceTree),
+        prompt: buildOperationalProfilePrompt(sourceTree, sourceSpans),
         schema: OperationalProfilePromptSchema,
         maxTokens: budget.maxTokens,
         taskKind: "extraction_operational_profile",