@agent-pattern-labs/leads-rig 0.1.7 → 0.1.8

package/docs/ARCHITECTURE.md

@@ -82,4 +82,8 @@ The package also carries the `@agent-pattern-labs/iso-*` helper ecosystem for tr
 - page visits
 - ingest requests
 
+Lead records preserve optional enrichment fields such as `linkedinUrl`, `phone`,
+`address`, `streetAddress`, `location`, `city`, `region`, `postalCode`, and
+`country` through normalization, deduplication, and ingest.
+
 The validator intentionally accepts the same defaults the local runtime normalizes, but it fails missing source evidence, invalid URL fields, invalid `emailType`, invalid `verificationStatus`, invalid confidence, person leads without email, and generic catch-all inboxes such as `info@`, `hello@`, or similar organizational aliases. Any email record must be a named `person` lead with a non-generic email and a human owner visible in the evidence; role inboxes, unknown-owner emails, blocked emails, unsupported emails, and unnamed person-like emails fail validation. Summaries expose `goodLeadCount`, which only counts those high-quality named human email records.

package/lib/leadharness-leads.mjs

@@ -162,7 +162,15 @@ export function normalizeLead(input, { now = new Date().toISOString() } = {}) {
   const verificationStatus = normalizeVerificationStatus(input.verificationStatus, emailType);
   const confidence = clampInteger(input.confidence, 0, 100, 0);
   const foundAt = normalizeDateTime(input.foundAt, now);
+  const linkedinUrl = normalizeLinkedInUrl(input.linkedinUrl);
   const phone = stringValue(input.phone);
+  const address = compactWhitespace(input.address);
+  const streetAddress = compactWhitespace(input.streetAddress);
+  const location = compactWhitespace(input.location);
+  const city = compactWhitespace(input.city);
+  const region = compactWhitespace(input.region);
+  const postalCode = compactWhitespace(input.postalCode);
+  const country = compactWhitespace(input.country);
   const socialUrls = normalizeUrlList(input.socialUrls);
   const contactUrls = normalizeUrlList(input.contactUrls);
 
@@ -187,7 +195,15 @@ export function normalizeLead(input, { now = new Date().toISOString() } = {}) {
     verificationStatus,
     confidence,
     warnings: normalizeWarnings(input.warnings),
+    linkedinUrl,
     phone,
+    address,
+    streetAddress,
+    location,
+    city,
+    region,
+    postalCode,
+    country,
     socialUrls,
     contactUrls,
     foundAt,
@@ -547,6 +563,15 @@ function normalizeLeadRecord(lead) {
     sources,
     evidence: sources[0]?.evidence || compactWhitespace(lead?.evidence),
     warnings: normalizeWarnings(lead?.warnings),
+    linkedinUrl: normalizeLinkedInUrl(lead?.linkedinUrl),
+    phone: stringValue(lead?.phone),
+    address: compactWhitespace(lead?.address),
+    streetAddress: compactWhitespace(lead?.streetAddress),
+    location: compactWhitespace(lead?.location),
+    city: compactWhitespace(lead?.city),
+    region: compactWhitespace(lead?.region),
+    postalCode: compactWhitespace(lead?.postalCode),
+    country: compactWhitespace(lead?.country),
     socialUrls: normalizeUrlList(lead?.socialUrls),
     contactUrls: normalizeUrlList(lead?.contactUrls),
   };
@@ -574,7 +599,15 @@ export function mergeLeadRecords(current, candidate) {
       ? secondary.verificationStatus
       : primary.verificationStatus,
     confidence: Math.max(primary.confidence || 0, secondary.confidence || 0),
+    linkedinUrl: primary.linkedinUrl || secondary.linkedinUrl,
     phone: primary.phone || secondary.phone,
+    address: longerString(primary.address, secondary.address),
+    streetAddress: primary.streetAddress || secondary.streetAddress,
+    location: longerString(primary.location, secondary.location),
+    city: primary.city || secondary.city,
+    region: primary.region || secondary.region,
+    postalCode: primary.postalCode || secondary.postalCode,
+    country: primary.country || secondary.country,
     socialUrls: normalizeUrlList([...(primary.socialUrls || []), ...(secondary.socialUrls || [])]),
     contactUrls: normalizeUrlList([...(primary.contactUrls || []), ...(secondary.contactUrls || [])]),
     warnings: normalizeWarnings([...(primary.warnings || []), ...(secondary.warnings || [])]),
@@ -842,6 +875,25 @@ function normalizeUrlList(value) {
   return urls;
 }
 
+function normalizeLinkedInUrl(value) {
+  let raw = stringValue(value);
+  if (!raw) return '';
+  if (!raw.includes('://')) raw = `https://${raw}`;
+
+  try {
+    const url = new URL(raw);
+    if (url.protocol !== 'http:' && url.protocol !== 'https:') return '';
+    const host = url.hostname.toLowerCase();
+    if (host !== 'linkedin.com' && !host.endsWith('.linkedin.com')) return '';
+
+    const path = url.pathname.replace(/\/+$/, '');
+    if (!/^\/(in|pub|company|school|showcase)\//i.test(path)) return '';
+    return `https://www.linkedin.com${path}`;
+  } catch {
+    return '';
+  }
+}
+
 function normalizeDateTime(value, fallback) {
   const raw = stringValue(value);
   if (!raw) return fallback;
@@ -876,6 +928,12 @@ function compactWhitespace(value) {
   return stringValue(value).replace(/\s+/g, ' ');
 }
 
+function longerString(left, right) {
+  const first = stringValue(left);
+  const second = stringValue(right);
+  return second.length > first.length ? second : first;
+}
+
 function displayCompany(domain) {
   const clean = normalizeDomain(domain);
   if (!clean) return '';

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@agent-pattern-labs/leads-rig",
-  "version": "0.1.7",
+  "version": "0.1.8",
   "description": "Agentic public-web lead discovery harness with portable ingest artifacts",
   "type": "module",
   "author": "Razroo",

package/templates/contracts.json

@@ -30,7 +30,15 @@
         },
         { "name": "confidence", "type": "integer", "required": true, "min": 0, "max": 100 },
         { "name": "warnings", "type": "json" },
+        { "name": "linkedinUrl", "type": "url" },
         { "name": "phone", "type": "string" },
+        { "name": "address", "type": "string" },
+        { "name": "streetAddress", "type": "string" },
+        { "name": "location", "type": "string" },
+        { "name": "city", "type": "string" },
+        { "name": "region", "type": "string" },
+        { "name": "postalCode", "type": "string" },
+        { "name": "country", "type": "string" },
         { "name": "socialUrls", "type": "json" },
         { "name": "contactUrls", "type": "json" },
         { "name": "foundAt", "type": "datetime" }

package/templates/lead-schema.json

@@ -74,10 +74,30 @@
           "items": { "type": "string" },
           "default": []
         },
+        "linkedinUrl": {
+          "type": "string",
+          "description": "Optional canonical public LinkedIn profile or organization URL."
+        },
         "phone": {
           "type": "string",
           "description": "Optional normalized phone number found near the public contact evidence."
         },
+        "address": {
+          "type": "string",
+          "description": "Optional complete public business address."
+        },
+        "streetAddress": {
+          "type": "string",
+          "description": "Optional street-address component."
+        },
+        "location": {
+          "type": "string",
+          "description": "Optional display location for the lead."
+        },
+        "city": { "type": "string" },
+        "region": { "type": "string" },
+        "postalCode": { "type": "string" },
+        "country": { "type": "string" },
         "socialUrls": {
           "type": "array",
           "items": { "type": "string" },