@ijonis/geo-lint 0.1.5 → 0.2.1

package/dist/index.cjs

@@ -191,7 +191,13 @@ var DEFAULT_CONFIG = {
       "Tabs",
       "Tab"
     ],
-    enabledContentTypes: ["blog"]
+    enabledContentTypes: ["blog"],
+    organizationSameAs: [],
+    servicePagePatterns: []
+  },
+  technical: {
+    feedUrls: [],
+    llmsTxtUrl: ""
   },
   i18n: {
     locales: ["de", "en"],
@@ -266,12 +272,18 @@ function mergeWithDefaults(user) {
       acronymAllowlist: user.geo?.acronymAllowlist ?? DEFAULT_CONFIG.geo.acronymAllowlist,
       vagueHeadings: user.geo?.vagueHeadings ?? DEFAULT_CONFIG.geo.vagueHeadings,
       genericAuthorNames: user.geo?.genericAuthorNames ?? DEFAULT_CONFIG.geo.genericAuthorNames,
-      allowedHtmlTags: user.geo?.allowedHtmlTags ?? DEFAULT_CONFIG.geo.allowedHtmlTags
+      allowedHtmlTags: user.geo?.allowedHtmlTags ?? DEFAULT_CONFIG.geo.allowedHtmlTags,
+      organizationSameAs: user.geo?.organizationSameAs ?? DEFAULT_CONFIG.geo.organizationSameAs,
+      servicePagePatterns: user.geo?.servicePagePatterns ?? DEFAULT_CONFIG.geo.servicePagePatterns
     },
     i18n: {
       locales: user.i18n?.locales ?? DEFAULT_CONFIG.i18n.locales,
       defaultLocale: user.i18n?.defaultLocale ?? DEFAULT_CONFIG.i18n.defaultLocale
     },
+    technical: {
+      feedUrls: user.technical?.feedUrls ?? DEFAULT_CONFIG.technical.feedUrls,
+      llmsTxtUrl: user.technical?.llmsTxtUrl ?? DEFAULT_CONFIG.technical.llmsTxtUrl
+    },
     rules: { ...DEFAULT_CONFIG.rules, ...user.rules ?? {} },
     thresholds: {
       title: { ...DEFAULT_CONFIG.thresholds.title, ...user.thresholds?.title ?? {} },
@@ -886,6 +898,68 @@ var duplicateRules = [
   duplicateDescription
 ];
 
+// src/utils/plaintext-structure.ts
+var MAX_HEADING_LENGTH = 80;
+var MIN_TABLE_ROWS = 2;
+function detectPlaintextHeadings(text) {
+  const lines = text.split("\n");
+  const headings = [];
+  for (let i = 0; i < lines.length; i++) {
+    const line = lines[i].trim();
+    if (!line || line.length > MAX_HEADING_LENGTH) continue;
+    const nextLine = lines[i + 1]?.trim() ?? "";
+    const isFollowedByBlank = i + 1 >= lines.length || nextLine === "";
+    if (!isFollowedByBlank) continue;
+    if (/[.,;:]$/.test(line)) continue;
+    const isTitleCase = /^[A-ZÄÖÜ]/.test(line) && line.split(/\s+/).length <= 12;
+    const isAllCaps = line === line.toUpperCase() && /[A-ZÄÖÜ]/.test(line) && line.length > 2;
+    const isQuestion = line.endsWith("?");
+    if (isTitleCase || isAllCaps || isQuestion) {
+      const level = isAllCaps || line.split(/\s+/).length <= 4 ? 2 : 3;
+      headings.push({ level, text: line, line: i + 1 });
+    }
+  }
+  return headings;
+}
+function detectPlaintextTable(text) {
+  const lines = text.split("\n").filter((l) => l.trim().length > 0);
+  const tabLines = lines.filter((l) => l.includes("	"));
+  if (tabLines.length >= MIN_TABLE_ROWS) {
+    const colCounts = tabLines.map((l) => l.split("	").length);
+    const consistent = colCounts.every(
+      (c) => c === colCounts[0] && c >= 2
+    );
+    if (consistent) return true;
+  }
+  const spaceSeparated = lines.filter((l) => /\S {3,}\S/.test(l));
+  if (spaceSeparated.length >= MIN_TABLE_ROWS + 1) {
+    return true;
+  }
+  return false;
+}
+function detectPlaintextList(text) {
+  const listPattern = /^[\s]*[•·–—]\s+|^[\s]*\w\)\s+|^[\s]*\d+\)\s+/m;
+  const lines = text.split("\n").filter((l) => listPattern.test(l));
+  return lines.length >= 2;
+}
+function detectPlaintextFaq(text) {
+  const lines = text.split("\n");
+  let questionCount = 0;
+  for (let i = 0; i < lines.length; i++) {
+    const line = lines[i].trim();
+    if (!line.endsWith("?")) continue;
+    if (line.length > MAX_HEADING_LENGTH) continue;
+    const nextContent = lines.slice(i + 1).find((l) => l.trim().length > 0);
+    if (nextContent && nextContent.trim().length > line.length) {
+      questionCount++;
+    }
+  }
+  return {
+    hasFaq: questionCount >= 2,
+    questionCount
+  };
+}
+
 // src/utils/heading-extractor.ts
 function isInCodeBlock(lines, lineIndex) {
   let inCodeBlock = false;
@@ -897,7 +971,7 @@ function isInCodeBlock(lines, lineIndex) {
   }
   return inCodeBlock;
 }
-function extractHeadings(mdxBody) {
+function extractHeadings(mdxBody, contentSource) {
   const headings = [];
   const lines = mdxBody.split("\n");
   const headingRegex = /^(#{1,6})\s+(.+)$/;
@@ -914,6 +988,9 @@ function extractHeadings(mdxBody) {
       });
     }
   }
+  if (headings.length === 0 && contentSource === "url") {
+    return detectPlaintextHeadings(mdxBody);
+  }
   return headings;
 }
 function countH1s(headings) {
@@ -945,6 +1022,9 @@ var missingH1 = {
   category: "seo",
   fixStrategy: "Add an H1 heading (# Heading) at the start of the content",
   run: (item) => {
+    if (item.contentSource === "url") {
+      return [];
+    }
     if (item.contentType === "blog") {
       return [];
     }
@@ -1231,8 +1311,16 @@ function countWords(text) {
 }
 function countSentences(text) {
   const stripped = stripMarkdown(text);
-  const sentences = stripped.match(/[.!?]+(?:\s|$)/g);
-  return sentences ? sentences.length : 0;
+  const sentenceEndings = stripped.match(/[.!?]+(?:\s|$|(?=[A-ZÄÖÜ]))/g);
+  if (sentenceEndings && sentenceEndings.length > 0) {
+    return sentenceEndings.length;
+  }
+  const lines = stripped.split(/\n+/).filter((l) => l.trim().length > 20);
+  if (lines.length > 1) {
+    return lines.length;
+  }
+  const hasWords = /\w{2,}/.test(stripped);
+  return hasWords ? 1 : 0;
 }
 
 // src/utils/readability.ts
@@ -1514,6 +1602,7 @@ var robotsRules = [
 // src/rules/slug-rules.ts
 var SLUG_DEFAULTS = { maxLength: 75 };
 var SLUG_PATTERN = /^[a-z0-9]+(?:-[a-z0-9]+)*$/;
+var URL_PATH_PATTERN = /^[a-z0-9]+(?:[-/][a-z0-9]+)*$/;
 var slugInvalidCharacters = {
   name: "slug-invalid-characters",
   severity: "error",
@@ -1521,8 +1610,10 @@ var slugInvalidCharacters = {
   fixStrategy: 'Use lowercase alphanumeric characters with hyphens only (e.g., "my-blog-post")',
   run: (item) => {
     if (!item.slug) return [];
+    const isUrl = item.contentSource === "url";
+    const pattern = isUrl ? URL_PATH_PATTERN : SLUG_PATTERN;
     const hasUppercase = /[A-Z]/.test(item.slug);
-    const matchesPattern = SLUG_PATTERN.test(item.slug);
+    const matchesPattern = pattern.test(item.slug);
     if (hasUppercase || !matchesPattern) {
       return [{
         file: getDisplayPath(item),
@@ -1530,7 +1621,7 @@ var slugInvalidCharacters = {
         rule: "slug-invalid-characters",
         severity: "error",
         message: `Slug "${item.slug}" contains invalid characters`,
-        suggestion: 'Slugs must be lowercase alphanumeric with hyphens only (e.g., "my-blog-post")'
+        suggestion: isUrl ? "URL paths must be lowercase alphanumeric with hyphens and slashes only" : 'Slugs must be lowercase alphanumeric with hyphens only (e.g., "my-blog-post")'
       }];
     }
     return [];
@@ -1795,8 +1886,8 @@ var WEAK_LEAD_STARTS = [
   "schauen wir uns"
 ];
 var TABLE_SEPARATOR_PATTERN = /\|\s*:?-{2,}/;
-function countQuestionHeadings(body) {
-  const headings = extractHeadings(body);
+function countQuestionHeadings(body, contentSource) {
+  const headings = extractHeadings(body, contentSource);
   let count = 0;
   for (const heading of headings) {
     const text = heading.text.trim();
@@ -1858,12 +1949,20 @@ function countStatistics(body) {
   }
   return matches.size;
 }
-function hasFAQSection(body) {
+function hasFAQSection(body, contentSource) {
   const faqPattern = /#{2,3}\s*(FAQ|Häufige Fragen|Frequently Asked|Fragen und Antworten)/i;
-  return faqPattern.test(body);
+  if (faqPattern.test(body)) return true;
+  if (contentSource === "url") {
+    return detectPlaintextFaq(body).hasFaq;
+  }
+  return false;
 }
-function hasMarkdownTable(body) {
-  return TABLE_SEPARATOR_PATTERN.test(body);
+function hasMarkdownTable(body, contentSource) {
+  if (TABLE_SEPARATOR_PATTERN.test(body)) return true;
+  if (contentSource === "url") {
+    return detectPlaintextTable(body);
+  }
+  return false;
 }
 function countEntityMentions(body, entity) {
   const escapedEntity = entity.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
@@ -2025,7 +2124,7 @@ function getParagraphs(body) {
   }
   return paragraphs;
 }
-function hasMarkdownList(body) {
+function hasMarkdownList(body, contentSource) {
   const lines = body.split("\n");
   let inCodeBlock = false;
   for (const line of lines) {
@@ -2038,6 +2137,9 @@ function hasMarkdownList(body) {
     if (/^[-*]\s+/.test(trimmed)) return true;
     if (/^\d+\.\s+/.test(trimmed)) return true;
   }
+  if (contentSource === "url") {
+    return detectPlaintextList(body);
+  }
   return false;
 }
 function countInternalLinks(body) {
@@ -2622,12 +2724,72 @@ var datasetSchemaReadiness = {
     return results;
   }
 };
-var schemaRules = [
+var MIN_SAMEAS_ENTRIES = 2;
+function createSchemaSameAsRule(organizationSameAs) {
+  let hasFired = false;
+  return {
+    name: "seo-schema-sameas-incomplete",
+    severity: "warning",
+    category: "seo",
+    fixStrategy: "Add social profiles (LinkedIn, GitHub, Twitter), Wikidata QID, and Crunchbase URL to Organization schema sameAs array",
+    run: (_item) => {
+      if (hasFired) return [];
+      hasFired = true;
+      if (!organizationSameAs || organizationSameAs.length === 0) return [];
+      if (organizationSameAs.length < MIN_SAMEAS_ENTRIES) {
+        return [
+          {
+            file: "_site",
+            field: "schema",
+            rule: "seo-schema-sameas-incomplete",
+            severity: "warning",
+            message: `Organization sameAs has ${organizationSameAs.length} entry \u2014 include at least ${MIN_SAMEAS_ENTRIES} for entity verification`,
+            suggestion: "AI models use sameAs to verify entity identity. Include at least LinkedIn + one other profile (GitHub, Wikidata QID, Crunchbase)."
+          }
+        ];
+      }
+      return [];
+    }
+  };
+}
+function createServicePageSchemaRule(servicePagePatterns) {
+  return {
+    name: "seo-service-page-no-schema",
+    severity: "warning",
+    category: "seo",
+    fixStrategy: "Add Service structured data (JSON-LD) to service pages with name, description, provider, and areaServed.",
+    run: (item) => {
+      if (!servicePagePatterns || servicePagePatterns.length === 0) return [];
+      const matchesPattern = servicePagePatterns.some(
+        (pattern) => item.permalink.includes(pattern)
+      );
+      if (!matchesPattern) return [];
+      return [
+        {
+          file: getDisplayPath(item),
+          field: "schema",
+          rule: "seo-service-page-no-schema",
+          severity: "warning",
+          message: `Service page "${item.permalink}" should have Service structured data`,
+          suggestion: 'Service pages need schema markup to appear in AI answers for "[service] provider in [city]" queries. Add Service JSON-LD with name, description, provider, and areaServed.'
+        }
+      ];
+    }
+  };
+}
+var schemaStaticRules = [
   blogMissingSchemaFields,
   faqpageSchemaReadiness,
   breadcrumblistSchemaReadiness,
   datasetSchemaReadiness
 ];
+function createSchemaRules(geo) {
+  return [
+    ...schemaStaticRules,
+    createSchemaSameAsRule(geo.organizationSameAs),
+    createServicePageSchemaRule(geo.servicePagePatterns)
+  ];
+}
 
 // src/rules/keyword-coherence-rules.ts
 var MIN_SIGNIFICANT_WORDS = 2;
@@ -13709,8 +13871,27 @@ function jaccardSimilarity(a, b) {
   const union = a.size + b.size - intersection;
   return union > 0 ? intersection / union : 0;
 }
+var REFERENCE_PATTERNS = [
+  /archived from the original on/gi,
+  /retrieved (?:on )?\d/gi,
+  /accessed (?:on )?\d/gi,
+  /cite (?:web|book|journal|news)/gi,
+  /\^\s*\[?\d+\]?/g,
+  /isbn \d/gi,
+  /doi:\s*\d/gi,
+  /pmid:\s*\d/gi
+];
+function stripReferenceBoilerplate(text) {
+  let result = text;
+  for (const pattern of REFERENCE_PATTERNS) {
+    result = result.replace(pattern, "");
+  }
+  result = result.replace(/\n(?:references|sources|bibliography|einzelnachweise|weblinks)\n[\s\S]*$/i, "");
+  return result;
+}
 function analyzeRepetition(body) {
-  const plain = stripMarkdown(body).toLowerCase();
+  const cleaned = stripReferenceBoilerplate(body);
+  const plain = stripMarkdown(cleaned).toLowerCase();
   const words = plain.replace(/[^\p{L}\p{N}\s]/gu, " ").split(/\s+/).filter((w) => w.length > 0);
   const fiveGrams = extractNgrams(words, 5);
   const phraseCounts = /* @__PURE__ */ new Map();
@@ -13719,7 +13900,7 @@ function analyzeRepetition(body) {
   }
   const repeatedPhrases = [...phraseCounts.entries()].filter(([, count]) => count >= 3).sort((a, b) => b[1] - a[1]);
   const topRepeatedPhrases = repeatedPhrases.slice(0, 5).map(([phrase, count]) => ({ phrase, count }));
-  const paragraphs = body.split(/\n\s*\n/).map((p) => p.trim()).filter((p) => p.length > 0 && !p.startsWith("#") && !p.startsWith("|"));
+  const paragraphs = cleaned.split(/\n\s*\n/).map((p) => p.trim()).filter((p) => p.length > 0 && !p.startsWith("#") && !p.startsWith("|"));
   let totalSimilarity = 0;
   let pairCount = 0;
   for (let i = 0; i < paragraphs.length; i++) {
@@ -14025,10 +14206,10 @@ var geoNoQuestionHeadings = {
     if (!geoTypes.includes(item.contentType)) return [];
     const wordCount = countWords(item.body);
     if (wordCount < GEO_MIN_WORDS) return [];
-    const headings = extractHeadings(item.body);
+    const headings = extractHeadings(item.body, item.contentSource);
     const subHeadings = headings.filter((h) => h.level === 2 || h.level === 3);
     if (subHeadings.length === 0) return [];
-    const questionCount = countQuestionHeadings(item.body);
+    const questionCount = countQuestionHeadings(item.body, item.contentSource);
     const ratio = questionCount / subHeadings.length;
     if (ratio < QUESTION_HEADING_THRESHOLD) {
       return [{
@@ -14118,7 +14299,7 @@ var geoMissingFaqSection = {
     if (!geoTypes.includes(item.contentType)) return [];
     const wordCount = countWords(item.body);
     if (wordCount < FAQ_MIN_WORDS) return [];
-    if (!hasFAQSection(item.body)) {
+    if (!hasFAQSection(item.body, item.contentSource)) {
       return [{
         file: getDisplayPath(item),
         field: "body",
@@ -14163,7 +14344,7 @@ var geoMissingTable = {
     if (!geoTypes.includes(item.contentType)) return [];
     const wordCount = countWords(item.body);
     if (wordCount < TABLE_MIN_WORDS) return [];
-    if (!hasMarkdownTable(item.body)) {
+    if (!hasMarkdownTable(item.body, item.contentSource)) {
       return [{
         file: getDisplayPath(item),
         field: "body",
@@ -14659,6 +14840,58 @@ var geoMissingTldr = {
     return [];
   }
 };
+var ORG_AUTHOR_PATTERNS = [
+  /\bteam\b/i,
+  /\bredaktion\b/i,
+  /\beditorial\b/i,
+  /\beditors?\b/i,
+  /\bherausgeber\b/i,
+  /\bverlag\b/i,
+  /\bredaktionsteam\b/i
+];
+function createGeoAuthorNotPersonRule(brandName) {
+  return {
+    name: "geo-author-not-person",
+    severity: "warning",
+    category: "geo",
+    fixStrategy: "Replace organization name with individual author name. Use Person type in BlogPosting schema for stronger E-E-A-T signals.",
+    run: (item, context) => {
+      const geoTypes = context.geoEnabledContentTypes ?? ["blog"];
+      if (!geoTypes.includes(item.contentType)) return [];
+      if (!item.author || item.author.trim() === "") return [];
+      if (!brandName || brandName.trim() === "") return [];
+      const normalizedAuthor = item.author.trim().toLowerCase();
+      if (normalizedAuthor === brandName.trim().toLowerCase()) {
+        return [
+          {
+            file: getDisplayPath(item),
+            field: "author",
+            rule: "geo-author-not-person",
+            severity: "warning",
+            message: `Author "${item.author}" is the organization name \u2014 use a person's name instead`,
+            suggestion: "AI models cite named experts over faceless organizations. Use the actual author's name for stronger E-E-A-T signals."
+          }
+        ];
+      }
+      const matchesOrgPattern = ORG_AUTHOR_PATTERNS.some(
+        (pattern) => pattern.test(item.author)
+      );
+      if (matchesOrgPattern) {
+        return [
+          {
+            file: getDisplayPath(item),
+            field: "author",
+            rule: "geo-author-not-person",
+            severity: "warning",
+            message: `Author "${item.author}" appears to be an organization or team name`,
+            suggestion: "BlogPosting with author.@type: Person gets cited more than Organization. Use an individual person's name."
+          }
+        ];
+      }
+      return [];
+    }
+  };
+}
 var geoEeatStaticRules = [
   geoMissingSourceCitations,
   geoMissingExpertQuotes,
@@ -14671,7 +14904,8 @@ function createGeoEeatRules(geo) {
   return [
     ...geoEeatStaticRules,
     createGeoMissingAuthorRule(geo.genericAuthorNames ?? []),
-    createGeoHeadingTooVagueRule(geo.vagueHeadings ?? [])
+    createGeoHeadingTooVagueRule(geo.vagueHeadings ?? []),
+    createGeoAuthorNotPersonRule(geo.brandName)
   ];
 }
 
@@ -14758,7 +14992,7 @@ var geoMissingLists = {
     if (!geoTypes.includes(item.contentType)) return [];
     const wordCount = countWords(item.body);
     if (wordCount < STRUCTURE_MIN_WORDS) return [];
-    if (!hasMarkdownList(item.body)) {
+    if (!hasMarkdownList(item.body, item.contentSource)) {
       return [{
         file: getDisplayPath(item),
         field: "body",
@@ -15660,6 +15894,68 @@ var contentQualityRules = [
   sentenceVariety
 ];
 
+// src/rules/technical-site-rules.ts
+function createNoFeedRule(feedUrls) {
+  let hasFired = false;
+  return {
+    name: "technical-no-feed",
+    severity: "warning",
+    category: "technical",
+    fixStrategy: "Add an RSS or JSON feed endpoint exposing blog posts with full content.",
+    run: (_item, _context) => {
+      if (hasFired) return [];
+      hasFired = true;
+      if (feedUrls === void 0) return [];
+      if (feedUrls.length === 0) {
+        return [
+          {
+            file: "_site",
+            field: "feed",
+            rule: "technical-no-feed",
+            severity: "warning",
+            message: "No RSS/Atom/JSON feed detected \u2014 AI systems lose a structured ingestion path",
+            suggestion: "Feeds provide a structured ingestion path for AI systems beyond crawler discovery. Add an RSS or JSON feed endpoint."
+          }
+        ];
+      }
+      return [];
+    }
+  };
+}
+function createNoLlmsTxtRule(llmsTxtUrl) {
+  let hasFired = false;
+  return {
+    name: "technical-no-llms-txt",
+    severity: "warning",
+    category: "technical",
+    fixStrategy: "Create a /llms.txt endpoint that maps your most important content for LLM consumption in Markdown format.",
+    run: (_item, _context) => {
+      if (hasFired) return [];
+      hasFired = true;
+      if (llmsTxtUrl === void 0) return [];
+      if (llmsTxtUrl.trim() === "") {
+        return [
+          {
+            file: "_site",
+            field: "llms-txt",
+            rule: "technical-no-llms-txt",
+            severity: "warning",
+            message: "No /llms.txt endpoint detected \u2014 missing the emerging standard for LLM content declaration",
+            suggestion: "llms.txt is the robots.txt equivalent for AI \u2014 trivial to add, future-proofs your site for LLM crawlers."
+          }
+        ];
+      }
+      return [];
+    }
+  };
+}
+function createTechnicalSiteRules(technical) {
+  return [
+    createNoFeedRule(technical.feedUrls),
+    createNoLlmsTxtRule(technical.llmsTxtUrl)
+  ];
+}
+
 // src/rules/index.ts
 function buildRules(config, linkExtractor) {
   const rules = [
@@ -15679,7 +15975,7 @@ function buildRules(config, linkExtractor) {
     ...createI18nRules(config.i18n),
     ...dateRules,
     ...config.categories.length > 0 ? createCategoryRules(config.categories) : [],
-    ...schemaRules,
+    ...createSchemaRules(config.geo),
     ...createGeoRules(config.geo),
     ...createGeoEeatRules(config.geo),
     ...geoStructureRules,
@@ -15687,7 +15983,8 @@ function buildRules(config, linkExtractor) {
     ...createGeoRagRules(config.geo),
     ...keywordCoherenceRules,
     ...createCanonicalRules(config.siteUrl),
-    ...contentQualityRules
+    ...contentQualityRules,
+    ...createTechnicalSiteRules(config.technical)
   ];
   return rules.map((rule) => applyRuleOverride(rule, config.rules));
 }