aiseo-audit 1.2.7 → 1.2.9

package/README.md

@@ -5,7 +5,7 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-7EB6D7.svg)](https://opensource.org/licenses/MIT)
 [![Node.js](https://img.shields.io/badge/node-%3E%3D20-7EB6D7.svg)](https://nodejs.org)
 [![TypeScript](https://img.shields.io/badge/TypeScript-5.9-7EB6D7?logo=typescript&logoColor=white)](https://www.typescriptlang.org/)
-[![Tests](https://img.shields.io/badge/tests-194%20passed-8FBC8F)](https://github.com/agencyenterprise/aiseo-audit)
+[![Tests](https://img.shields.io/badge/tests-202%20passed-8FBC8F)](https://github.com/agencyenterprise/aiseo-audit)
 [![Coverage](https://img.shields.io/codecov/c/github/agencyenterprise/aiseo-audit?color=8FBC8F&label=coverage)](https://codecov.io/gh/agencyenterprise/aiseo-audit)
 
 <div align="center">
@@ -15,6 +15,9 @@
 
 Deterministic CLI that audits web pages for **AI search readiness**. Think Lighthouse, but for how well AI engines can fetch, extract, understand, and cite your content.
 
+> [!TIP]
+> Run `aiseo-audit https://www.aiseo-audit.com` to see an [A+ Score](https://www.aiseo-audit.com/)
+
 **AI SEO measures how reusable your content is for generative engines, not traditional search rankings.**
 
 - [Quick Start](#quick-start)

package/dist/cli.js

@@ -27,7 +27,7 @@ var import_commander = require("commander");
 
 // src/modules/analyzer/constants.ts
 var DOMAIN_SIGNAL_TIMEOUT_CAP = 5e3;
-var VERSION = true ? "1.2.7" : "0.0.0";
+var VERSION = true ? "1.2.9" : "0.0.0";
 
 // src/modules/fetcher/constants.ts
 var MAX_RESPONSE_SIZE = 10 * 1024 * 1024;
@@ -851,7 +851,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const extractRatio = page.stats.rawByteLength > 0 ? page.stats.cleanTextLength / page.stats.rawByteLength : 0;
-  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio >= 0.01 ? 8 : extractRatio > 0.15 ? 10 : 2;
+  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio > 0.15 ? 10 : extractRatio >= 0.01 ? 8 : 2;
   factors.push(
     makeFactor(
       "Text Extraction Quality",
@@ -876,7 +876,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const wc = page.stats.wordCount;
-  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc >= 100 ? 8 : wc > 3e3 ? 10 : 2;
+  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc > 3e3 ? 10 : wc >= 100 ? 8 : 2;
   factors.push(makeFactor("Word Count Adequacy", wcScore, 12, `${wc} words`));
   if (domainSignals) {
     const access2 = checkCrawlerAccess(domainSignals.robotsTxt);
@@ -1359,7 +1359,11 @@ function removeBoilerplate($) {
 function normalizeWhitespace(text) {
   return text.replace(/\s+/g, " ").trim();
 }
+var BLOCK_ELEMENTS = "p,div,td,th,li,h1,h2,h3,h4,h5,h6,dt,dd,br,blockquote,section,article";
 function extractCleanText($) {
+  $(BLOCK_ELEMENTS).each((_, el) => {
+    $(el).append(" ");
+  });
   return normalizeWhitespace($("body").text());
 }
 
@@ -1368,7 +1372,9 @@ function extractPage(html, url) {
   const $ = cheerio.load(html);
   const title = $("title").text().trim() || $('meta[property="og:title"]').attr("content")?.trim() || "";
   const metaDescription = $('meta[name="description"]').attr("content")?.trim() || $('meta[property="og:description"]').attr("content")?.trim() || "";
-  const rawText = $("body").text().replace(/\s+/g, " ").trim();
+  const $raw = cheerio.load(html);
+  $raw("script, style, noscript").remove();
+  const rawText = $raw("body").text().replace(/\s+/g, " ").trim();
   const rawByteLength = Buffer.byteLength(html, "utf-8");
   const h1Count = $("h1").length;
   const h2Count = $("h2").length;

package/dist/cli.mjs

@@ -3,7 +3,7 @@ import { Command } from "commander";
 
 // src/modules/analyzer/constants.ts
 var DOMAIN_SIGNAL_TIMEOUT_CAP = 5e3;
-var VERSION = true ? "1.2.7" : "0.0.0";
+var VERSION = true ? "1.2.9" : "0.0.0";
 
 // src/modules/fetcher/constants.ts
 var MAX_RESPONSE_SIZE = 10 * 1024 * 1024;
@@ -827,7 +827,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const extractRatio = page.stats.rawByteLength > 0 ? page.stats.cleanTextLength / page.stats.rawByteLength : 0;
-  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio >= 0.01 ? 8 : extractRatio > 0.15 ? 10 : 2;
+  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio > 0.15 ? 10 : extractRatio >= 0.01 ? 8 : 2;
   factors.push(
     makeFactor(
       "Text Extraction Quality",
@@ -852,7 +852,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const wc = page.stats.wordCount;
-  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc >= 100 ? 8 : wc > 3e3 ? 10 : 2;
+  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc > 3e3 ? 10 : wc >= 100 ? 8 : 2;
   factors.push(makeFactor("Word Count Adequacy", wcScore, 12, `${wc} words`));
   if (domainSignals) {
     const access2 = checkCrawlerAccess(domainSignals.robotsTxt);
@@ -1335,7 +1335,11 @@ function removeBoilerplate($) {
 function normalizeWhitespace(text) {
   return text.replace(/\s+/g, " ").trim();
 }
+var BLOCK_ELEMENTS = "p,div,td,th,li,h1,h2,h3,h4,h5,h6,dt,dd,br,blockquote,section,article";
 function extractCleanText($) {
+  $(BLOCK_ELEMENTS).each((_, el) => {
+    $(el).append(" ");
+  });
   return normalizeWhitespace($("body").text());
 }
 
@@ -1344,7 +1348,9 @@ function extractPage(html, url) {
   const $ = cheerio.load(html);
   const title = $("title").text().trim() || $('meta[property="og:title"]').attr("content")?.trim() || "";
   const metaDescription = $('meta[name="description"]').attr("content")?.trim() || $('meta[property="og:description"]').attr("content")?.trim() || "";
-  const rawText = $("body").text().replace(/\s+/g, " ").trim();
+  const $raw = cheerio.load(html);
+  $raw("script, style, noscript").remove();
+  const rawText = $raw("body").text().replace(/\s+/g, " ").trim();
   const rawByteLength = Buffer.byteLength(html, "utf-8");
   const h1Count = $("h1").length;
   const h2Count = $("h2").length;

package/dist/index.js

@@ -849,7 +849,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const extractRatio = page.stats.rawByteLength > 0 ? page.stats.cleanTextLength / page.stats.rawByteLength : 0;
-  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio >= 0.01 ? 8 : extractRatio > 0.15 ? 10 : 2;
+  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio > 0.15 ? 10 : extractRatio >= 0.01 ? 8 : 2;
   factors.push(
     makeFactor(
       "Text Extraction Quality",
@@ -874,7 +874,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const wc = page.stats.wordCount;
-  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc >= 100 ? 8 : wc > 3e3 ? 10 : 2;
+  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc > 3e3 ? 10 : wc >= 100 ? 8 : 2;
   factors.push(makeFactor("Word Count Adequacy", wcScore, 12, `${wc} words`));
   if (domainSignals) {
     const access2 = checkCrawlerAccess(domainSignals.robotsTxt);
@@ -1357,7 +1357,11 @@ function removeBoilerplate($) {
 function normalizeWhitespace(text) {
   return text.replace(/\s+/g, " ").trim();
 }
+var BLOCK_ELEMENTS = "p,div,td,th,li,h1,h2,h3,h4,h5,h6,dt,dd,br,blockquote,section,article";
 function extractCleanText($) {
+  $(BLOCK_ELEMENTS).each((_, el) => {
+    $(el).append(" ");
+  });
   return normalizeWhitespace($("body").text());
 }
 
@@ -1366,7 +1370,9 @@ function extractPage(html, url) {
   const $ = cheerio.load(html);
   const title = $("title").text().trim() || $('meta[property="og:title"]').attr("content")?.trim() || "";
   const metaDescription = $('meta[name="description"]').attr("content")?.trim() || $('meta[property="og:description"]').attr("content")?.trim() || "";
-  const rawText = $("body").text().replace(/\s+/g, " ").trim();
+  const $raw = cheerio.load(html);
+  $raw("script, style, noscript").remove();
+  const rawText = $raw("body").text().replace(/\s+/g, " ").trim();
   const rawByteLength = Buffer.byteLength(html, "utf-8");
   const h1Count = $("h1").length;
   const h2Count = $("h2").length;
@@ -1440,7 +1446,7 @@ var import_zod = require("zod");
 
 // src/modules/analyzer/constants.ts
 var DOMAIN_SIGNAL_TIMEOUT_CAP = 5e3;
-var VERSION = true ? "1.2.7" : "0.0.0";
+var VERSION = true ? "1.2.9" : "0.0.0";
 
 // src/modules/fetcher/schema.ts
 var FetchOptionsSchema = import_zod.z.object({

package/dist/index.mjs

@@ -811,7 +811,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const extractRatio = page.stats.rawByteLength > 0 ? page.stats.cleanTextLength / page.stats.rawByteLength : 0;
-  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio >= 0.01 ? 8 : extractRatio > 0.15 ? 10 : 2;
+  const extractScore = extractRatio >= 0.05 && extractRatio <= 0.15 ? 12 : extractRatio > 0.15 ? 10 : extractRatio >= 0.01 ? 8 : 2;
   factors.push(
     makeFactor(
       "Text Extraction Quality",
@@ -836,7 +836,7 @@ function auditContentExtractability(page, fetchResult, domainSignals) {
     )
   );
   const wc = page.stats.wordCount;
-  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc >= 100 ? 8 : wc > 3e3 ? 10 : 2;
+  const wcScore = wc >= 300 && wc <= 3e3 ? 12 : wc > 3e3 ? 10 : wc >= 100 ? 8 : 2;
   factors.push(makeFactor("Word Count Adequacy", wcScore, 12, `${wc} words`));
   if (domainSignals) {
     const access2 = checkCrawlerAccess(domainSignals.robotsTxt);
@@ -1319,7 +1319,11 @@ function removeBoilerplate($) {
 function normalizeWhitespace(text) {
   return text.replace(/\s+/g, " ").trim();
 }
+var BLOCK_ELEMENTS = "p,div,td,th,li,h1,h2,h3,h4,h5,h6,dt,dd,br,blockquote,section,article";
 function extractCleanText($) {
+  $(BLOCK_ELEMENTS).each((_, el) => {
+    $(el).append(" ");
+  });
   return normalizeWhitespace($("body").text());
 }
 
@@ -1328,7 +1332,9 @@ function extractPage(html, url) {
   const $ = cheerio.load(html);
   const title = $("title").text().trim() || $('meta[property="og:title"]').attr("content")?.trim() || "";
   const metaDescription = $('meta[name="description"]').attr("content")?.trim() || $('meta[property="og:description"]').attr("content")?.trim() || "";
-  const rawText = $("body").text().replace(/\s+/g, " ").trim();
+  const $raw = cheerio.load(html);
+  $raw("script, style, noscript").remove();
+  const rawText = $raw("body").text().replace(/\s+/g, " ").trim();
   const rawByteLength = Buffer.byteLength(html, "utf-8");
   const h1Count = $("h1").length;
   const h2Count = $("h2").length;
@@ -1402,7 +1408,7 @@ import { z } from "zod";
 
 // src/modules/analyzer/constants.ts
 var DOMAIN_SIGNAL_TIMEOUT_CAP = 5e3;
-var VERSION = true ? "1.2.7" : "0.0.0";
+var VERSION = true ? "1.2.9" : "0.0.0";
 
 // src/modules/fetcher/schema.ts
 var FetchOptionsSchema = z.object({

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "aiseo-audit",
-  "version": "1.2.7",
+  "version": "1.2.9",
   "description": "Lighthouse for AI SEO. Audit any webpage for AI search readiness. 7 categories, 30+ factors, research-backed scoring. Deterministic, engine-agnostic, zero API keys.",
   "main": "./dist/index.js",
   "module": "./dist/index.mjs",