pagesight 0.15.0 → 0.16.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pagesight",
-  "version": "0.15.0",
+  "version": "0.16.0",
   "description": "See your site the way search engines and AI see it.",
   "keywords": [
     "seo",

package/src/tools/search.ts

@@ -466,15 +466,103 @@ function daysAgo(n: number): string {
   return parts; // en-CA formats as YYYY-MM-DD
 }
 
+// ── Keyword gap analysis ──
+
+function extractPageText(html: string): string {
+  // Strip scripts, styles, and HTML tags to get visible text content
+  let text = html;
+  text = text.replace(/<script[^>]*>[\s\S]*?<\/script>/gi, " ");
+  text = text.replace(/<style[^>]*>[\s\S]*?<\/style>/gi, " ");
+  text = text.replace(/<[^>]+>/g, " ");
+  // Decode common HTML entities
+  text = text
+    .replace(/&amp;/g, "&")
+    .replace(/&lt;/g, "<")
+    .replace(/&gt;/g, ">")
+    .replace(/&quot;/g, '"')
+    .replace(/&#39;/g, "'")
+    .replace(/&nbsp;/g, " ");
+  // Normalize whitespace
+  text = text.replace(/\s+/g, " ").trim().toLowerCase();
+  return text;
+}
+
+function formatGapAnalysis(
+  pageUrl: string,
+  siteUrl: string,
+  queries: Array<{ query: string; clicks: number; impressions: number; ctr: number; position: number }>,
+  pageText: string,
+): string {
+  const lines: string[] = [`=== Keyword Gap Analysis: ${pageUrl} ===`, `Property: ${siteUrl}`, ""];
+
+  if (queries.length === 0) {
+    lines.push("No search queries found for this page.");
+    return lines.join("\n");
+  }
+
+  const onPage: typeof queries = [];
+  const gaps: typeof queries = [];
+
+  for (const q of queries) {
+    const words = q.query.toLowerCase().split(/\s+/);
+    const allPresent = words.every((w) => pageText.includes(w));
+    if (allPresent) {
+      onPage.push(q);
+    } else {
+      gaps.push(q);
+    }
+  }
+
+  lines.push(`Queries analyzed: ${queries.length}`);
+  lines.push(`On page: ${onPage.length}`);
+  lines.push(`Gaps (not on page): ${gaps.length}`);
+  lines.push("");
+
+  if (gaps.length > 0) {
+    lines.push("--- Keyword Gaps (queries with impressions but missing from page) ---", "");
+    gaps.sort((a, b) => b.impressions - a.impressions);
+    for (const q of gaps.slice(0, 20)) {
+      const ctr = (q.ctr * 100).toFixed(1);
+      lines.push(
+        `  "${q.query}" — ${q.impressions} imp, ${q.clicks} clicks, CTR ${ctr}%, pos ${q.position.toFixed(1)}`,
+      );
+    }
+    if (gaps.length > 20) lines.push(`  ... and ${gaps.length - 20} more`);
+    lines.push("");
+  }
+
+  if (onPage.length > 0) {
+    lines.push("--- Already Targeted (query found on page) ---", "");
+    onPage.sort((a, b) => b.impressions - a.impressions);
+    for (const q of onPage.slice(0, 10)) {
+      const ctr = (q.ctr * 100).toFixed(1);
+      lines.push(
+        `  "${q.query}" — ${q.impressions} imp, ${q.clicks} clicks, CTR ${ctr}%, pos ${q.position.toFixed(1)}`,
+      );
+    }
+    if (onPage.length > 10) lines.push(`  ... and ${onPage.length - 10} more`);
+    lines.push("");
+  }
+
+  const totalGapImpressions = gaps.reduce((sum, q) => sum + q.impressions, 0);
+  const totalImpressions = queries.reduce((sum, q) => sum + q.impressions, 0);
+  if (totalImpressions > 0) {
+    const gapPct = Math.round((totalGapImpressions / totalImpressions) * 100);
+    lines.push(`${gapPct}% of impressions come from queries not on the page — potential content opportunity.`);
+  }
+
+  return lines.join("\n");
+}
+
 // ── Tool registration ──
 
 export function registerSearchTool(server: McpServer): void {
   server.tool(
     "search",
-    "Query Google Search Console. Inspect URL indexing status, sample-inspect sitemap URLs, list properties and sitemaps, or analyze search traffic (clicks, impressions, CTR, rankings).",
+    "Query Google Search Console. Inspect URL indexing status, sample-inspect sitemap URLs, list properties and sitemaps, analyze search traffic, or find keyword gaps (queries with impressions but missing from page content).",
     {
       action: z
-        .enum(["inspect", "sample", "sitemaps", "analytics", "list_sites", "get_site", "get_sitemap"])
+        .enum(["inspect", "sample", "sitemaps", "analytics", "gaps", "list_sites", "get_site", "get_sitemap"])
         .optional()
         .describe("Action to perform. Auto-detected from params when unambiguous."),
       site_url: z.string().optional().describe("GSC property (e.g., 'sc-domain:example.com')."),
@@ -710,6 +798,46 @@ export function registerSearchTool(server: McpServer): void {
           return textResult(formatPerformance(site_url, result, dims, startDate, endDate));
         }
 
+        // ── gaps ──
+        if (resolvedAction === "gaps") {
+          if (!url) return textResult("Error: url is required for gap analysis.");
+          if (!site_url) return textResult("Error: site_url is required for gap analysis.");
+
+          // Fetch page content and GSC queries in parallel
+          const [pageRes, analyticsRes] = await Promise.all([
+            fetch(url, {
+              headers: { "User-Agent": "Mozilla/5.0 (compatible; Googlebot/2.1)", Accept: "text/html" },
+              redirect: "follow",
+            }),
+            querySearchAnalytics(site_url, {
+              startDate: start_date ?? daysAgo(28),
+              endDate: end_date ?? daysAgo(3),
+              dimensions: ["query"],
+              rowLimit: row_limit ?? 500,
+              dimensionFilterGroups: [
+                {
+                  groupType: "and" as const,
+                  filters: [{ dimension: "page", operator: "equals", expression: url } as SearchAnalyticsFilter],
+                },
+              ],
+            }),
+          ]);
+
+          if (!pageRes.ok) return textResult(`Error fetching ${url}: HTTP ${pageRes.status}`);
+          const html = await pageRes.text();
+          const pageText = extractPageText(html);
+
+          const queries = (analyticsRes.rows ?? []).map((row) => ({
+            query: row.keys[0],
+            clicks: row.clicks,
+            impressions: row.impressions,
+            ctr: row.ctr,
+            position: row.position,
+          }));
+
+          return textResult(formatGapAnalysis(url, site_url, queries, pageText));
+        }
+
         return textResult(`Error: Unknown action "${resolvedAction}".`);
       } catch (err) {
         const msg = err instanceof Error ? err.message : String(err);

package/src/tools/speed.ts

@@ -457,7 +457,8 @@ function formatBatchTable(results: Array<{ url: string; result: PsiResult }>, st
   if (sharedOpps.length > 0) {
     lines.push("--- Shared Opportunities ---", "");
     for (const [, { title, count, maxSavings }] of sharedOpps.slice(0, 10)) {
-      const savingsStr = maxSavings ? `, up to ${maxSavings}` : "";
+      const cleaned = maxSavings.replace(/^Est savings of /i, "");
+      const savingsStr = cleaned ? `, up to ${cleaned}` : "";
       lines.push(`  ${title} (${count}/${results.length} pages${savingsStr})`);
     }
     lines.push("");