@pseolint/core 0.2.2 → 0.3.0

package/README.md

@@ -24,15 +24,16 @@ console.log(`Findings: ${summary.findings.length}`);
 
 ## What It Checks
 
-37+ rules across 7 categories:
+42 rules across 8 categories. Seven categories feed the composite score; `data/*` is a separate data-binding family.
 
-- **Spam / SpamBrain risk** — near-duplicate (SimHash), entity-swap doorways, thin content, boilerplate ratio, template diversity, template coverage, publication velocity, doorway pattern
-- **Content** — unique value, heading / meta uniqueness, author attribution, E-E-A-T signals
-- **Internal linking** — orphan pages, dead ends, cluster connectivity, hub pages, link depth, unreachable-from-root
-- **Technical SEO** — canonical consistency, canonical/noindex and robots/noindex conflicts, sitemap completeness, robots compliance, redirect chains, soft 404s, Open Graph, hreflang
-- **Structured data** — JSON-LD validity, required fields, cross-page schema consistency
-- **Cannibalization** — title overlap, keyword collision, URL pattern conflicts
-- **Data binding** — verify rendered pages expose values from a source dataset (catches missing or identical-across-pages bindings)
+- **Spam / SpamBrain risk** (8) — near-duplicate (SimHash), entity-swap doorways, thin content, boilerplate ratio, template diversity, template coverage, publication velocity, doorway pattern
+- **Technical SEO** (8) — canonical consistency, canonical/noindex and robots/noindex conflicts, sitemap completeness, robots compliance, redirect chains, soft 404s, Open Graph, hreflang
+- **AEO / AI Overview citability** (8, v0.3.0) — `llms.txt` presence, AI-crawler access in robots.txt, freshness signals, FAQ coverage, answer-first opener, citable-fact density, non-replicable value, content modularity
+- **Content** (5) — unique value, heading / meta uniqueness, author attribution, E-E-A-T signals
+- **Internal linking** (5) — orphan pages, dead ends, cluster connectivity, hub pages, link depth
+- **Structured data** (3) — JSON-LD validity, required fields, cross-page schema consistency
+- **Cannibalization** (3) — title overlap, keyword collision, URL pattern conflicts
+- **Data binding** (2) — verify rendered pages expose values from a source dataset (missing or identical-across-pages bindings)
 
 ## API
 
@@ -118,4 +119,4 @@ All AI providers and `playwright-core` are optional peers — you only install t
 
 ## License
 
-MIT
+MIT

package/dist/ai/prompt.d.ts

@@ -1,5 +1,5 @@
 import type { RuleResult } from "../types.js";
-export declare const PROMPT_VERSION = "1.0.0";
+export declare const PROMPT_VERSION = "1.1.0";
 export declare const MAX_FINDINGS_IN_PROMPT = 200;
 export interface PromptRequest {
     system: string;

package/dist/ai/prompt.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"prompt.d.ts","sourceRoot":"","sources":["../../src/ai/prompt.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,UAAU,EAAY,MAAM,aAAa,CAAC;AAExD,eAAO,MAAM,cAAc,UAAU,CAAC;AACtC,eAAO,MAAM,sBAAsB,MAAM,CAAC;AAe1C,MAAM,WAAW,aAAa;IAC5B,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,MAAM,CAAC;CACd;AAED,wBAAgB,eAAe,CAAC,CAAC,EAAE,UAAU,GAAG,MAAM,CAMrD;AAmBD,wBAAgB,kBAAkB,CAAC,QAAQ,EAAE,UAAU,EAAE,EAAE,SAAS,EAAE,MAAM,GAAG,aAAa,CA8B3F"}
+{"version":3,"file":"prompt.d.ts","sourceRoot":"","sources":["../../src/ai/prompt.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,UAAU,EAAY,MAAM,aAAa,CAAC;AAExD,eAAO,MAAM,cAAc,UAAU,CAAC;AACtC,eAAO,MAAM,sBAAsB,MAAM,CAAC;AAqB1C,MAAM,WAAW,aAAa;IAC5B,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,MAAM,CAAC;CACd;AAED,wBAAgB,eAAe,CAAC,CAAC,EAAE,UAAU,GAAG,MAAM,CAMrD;AAqBD,wBAAgB,kBAAkB,CAAC,QAAQ,EAAE,UAAU,EAAE,EAAE,SAAS,EAAE,MAAM,GAAG,aAAa,CAqC3F"}

package/dist/ai/prompt.js

@@ -1,9 +1,15 @@
 import { createHash } from "node:crypto";
-export const PROMPT_VERSION = "1.0.0";
+export const PROMPT_VERSION = "1.1.0";
 export const MAX_FINDINGS_IN_PROMPT = 200;
 const SEVERITY_ORDER = { info: 0, warning: 1, error: 2, critical: 3 };
 const SYSTEM_PROMPT = `You are an SEO audit triage assistant. Given a list of pSEO linter findings, identify 1-5 underlying ROOT CAUSES driving the findings. Group findings by shared underlying problem, not by rule ID. Rank causes by likely SEO impact (highest first).
 
+Findings fall into two distinct threat families — treat them as separate root causes, not one combined cause:
+- SpamBrain penalty risk: spam/*, cannibal/*, content/*, data/*, tech/*, schema/*, links/* — these make Google penalize or demote the site.
+- AI Overview invisibility: aeo/* — these make pages uncitable in AI answer engines (ChatGPT, Perplexity, Gemini, AI Overviews). Sites not cited lose ~68% of traffic vs ~12% for cited sites.
+
+When both families are present, produce at least one root cause from each. Label AEO root causes clearly (e.g. "AI Overviews: ...") so the user can tell them apart from penalty risks.
+
 Rules:
 - Emit rootCauses FIRST, then narrative — do not reverse this order.
 - Keep each rootCause label <= 80 chars and phrase it as a problem statement.
@@ -31,11 +37,17 @@ export function buildPromptRequest(findings, pageCount) {
         pageUrl: f.pageUrl,
         group: f.group,
     }));
+    const countByCategory = {};
+    for (const f of findings) {
+        const cat = f.ruleId.split("/")[0];
+        countByCategory[cat] = (countByCategory[cat] ?? 0) + 1;
+    }
     const payload = {
         totalFindings: total,
         pageCount,
         truncated,
         findings: projected,
+        findingCountByCategory: countByCategory,
     };
     if (truncated) {
         const counts = {};

package/dist/ai/prompt.js.map

@@ -1 +1 @@
-{"version":3,"file":"prompt.js","sourceRoot":"","sources":["../../src/ai/prompt.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAC;AAGzC,MAAM,CAAC,MAAM,cAAc,GAAG,OAAO,CAAC;AACtC,MAAM,CAAC,MAAM,sBAAsB,GAAG,GAAG,CAAC;AAE1C,MAAM,cAAc,GAA6B,EAAE,IAAI,EAAE,CAAC,EAAE,OAAO,EAAE,CAAC,EAAE,KAAK,EAAE,CAAC,EAAE,QAAQ,EAAE,CAAC,EAAE,CAAC;AAEhG,MAAM,aAAa,GAAG;;;;;;;;;sDASgC,CAAC;AAOvD,MAAM,UAAU,eAAe,CAAC,CAAa;IAC3C,MAAM,IAAI,GAAG,UAAU,CAAC,QAAQ,CAAC;SAC9B,MAAM,CAAC,CAAC,CAAC,CAAC,OAAO,IAAI,EAAE,CAAC,GAAG,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC;SAC3C,MAAM,CAAC,KAAK,CAAC;SACb,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;IACf,OAAO,GAAG,CAAC,CAAC,MAAM,IAAI,IAAI,EAAE,CAAC;AAC/B,CAAC;AAmBD,MAAM,UAAU,kBAAkB,CAAC,QAAsB,EAAE,SAAiB;IAC1E,MAAM,KAAK,GAAG,QAAQ,CAAC,MAAM,CAAC;IAC9B,MAAM,MAAM,GAAG,CAAC,GAAG,QAAQ,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,cAAc,CAAC,CAAC,CAAC,QAAQ,CAAC,GAAG,cAAc,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC;IACrG,MAAM,SAAS,GAAG,KAAK,GAAG,sBAAsB,CAAC;IACjD,MAAM,SAAS,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,EAAE,sBAAsB,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAqB,EAAE,CAAC,CAAC;QACvF,EAAE,EAAE,eAAe,CAAC,CAAC,CAAC;QACtB,MAAM,EAAE,CAAC,CAAC,MAAM;QAChB,QAAQ,EAAE,CAAC,CAAC,QAAQ;QACpB,OAAO,EAAE,CAAC,CAAC,OAAO;QAClB,OAAO,EAAE,CAAC,CAAC,OAAO;QAClB,KAAK,EAAE,CAAC,CAAC,KAAK;KACf,CAAC,CAAC,CAAC;IAEJ,MAAM,OAAO,GAAkB;QAC7B,aAAa,EAAE,KAAK;QACpB,SAAS;QACT,SAAS;QACT,QAAQ,EAAE,SAAS;KACpB,CAAC;IAEF,IAAI,SAAS,EAAE,CAAC;QACd,MAAM,MAAM,GAA2B,EAAE,CAAC;QAC1C,KAAK,MAAM,CAAC,IAAI,QAAQ;YAAE,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC;QACzE,OAAO,CAAC,kBAAkB,GAAG,MAAM,CAAC;IACtC,CAAC;IAED,OAAO;QACL,MAAM,EAAE,aAAa;QACrB,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,OAAO,CAAC;KAC9B,CAAC;AACJ,CAAC"}
+{"version":3,"file":"prompt.js","sourceRoot":"","sources":["../../src/ai/prompt.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,EAAE,MAAM,aAAa,CAAC;AAGzC,MAAM,CAAC,MAAM,cAAc,GAAG,OAAO,CAAC;AACtC,MAAM,CAAC,MAAM,sBAAsB,GAAG,GAAG,CAAC;AAE1C,MAAM,cAAc,GAA6B,EAAE,IAAI,EAAE,CAAC,EAAE,OAAO,EAAE,CAAC,EAAE,KAAK,EAAE,CAAC,EAAE,QAAQ,EAAE,CAAC,EAAE,CAAC;AAEhG,MAAM,aAAa,GAAG;;;;;;;;;;;;;;;sDAegC,CAAC;AAOvD,MAAM,UAAU,eAAe,CAAC,CAAa;IAC3C,MAAM,IAAI,GAAG,UAAU,CAAC,QAAQ,CAAC;SAC9B,MAAM,CAAC,CAAC,CAAC,CAAC,OAAO,IAAI,EAAE,CAAC,GAAG,GAAG,GAAG,CAAC,CAAC,OAAO,CAAC;SAC3C,MAAM,CAAC,KAAK,CAAC;SACb,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;IACf,OAAO,GAAG,CAAC,CAAC,MAAM,IAAI,IAAI,EAAE,CAAC;AAC/B,CAAC;AAqBD,MAAM,UAAU,kBAAkB,CAAC,QAAsB,EAAE,SAAiB;IAC1E,MAAM,KAAK,GAAG,QAAQ,CAAC,MAAM,CAAC;IAC9B,MAAM,MAAM,GAAG,CAAC,GAAG,QAAQ,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,cAAc,CAAC,CAAC,CAAC,QAAQ,CAAC,GAAG,cAAc,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC;IACrG,MAAM,SAAS,GAAG,KAAK,GAAG,sBAAsB,CAAC;IACjD,MAAM,SAAS,GAAG,MAAM,CAAC,KAAK,CAAC,CAAC,EAAE,sBAAsB,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAqB,EAAE,CAAC,CAAC;QACvF,EAAE,EAAE,eAAe,CAAC,CAAC,CAAC;QACtB,MAAM,EAAE,CAAC,CAAC,MAAM;QAChB,QAAQ,EAAE,CAAC,CAAC,QAAQ;QACpB,OAAO,EAAE,CAAC,CAAC,OAAO;QAClB,OAAO,EAAE,CAAC,CAAC,OAAO;QAClB,KAAK,EAAE,CAAC,CAAC,KAAK;KACf,CAAC,CAAC,CAAC;IAEJ,MAAM,eAAe,GAA2B,EAAE,CAAC;IACnD,KAAK,MAAM,CAAC,IAAI,QAAQ,EAAE,CAAC;QACzB,MAAM,GAAG,GAAG,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,CAAC;QACnC,eAAe,CAAC,GAAG,CAAC,GAAG,CAAC,eAAe,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC;IACzD,CAAC;IAED,MAAM,OAAO,GAAkB;QAC7B,aAAa,EAAE,KAAK;QACpB,SAAS;QACT,SAAS;QACT,QAAQ,EAAE,SAAS;QACnB,sBAAsB,EAAE,eAAe;KACxC,CAAC;IAEF,IAAI,SAAS,EAAE,CAAC;QACd,MAAM,MAAM,GAA2B,EAAE,CAAC;QAC1C,KAAK,MAAM,CAAC,IAAI,QAAQ;YAAE,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC;QACzE,OAAO,CAAC,kBAAkB,GAAG,MAAM,CAAC;IACtC,CAAC;IAED,OAAO;QACL,MAAM,EAAE,aAAa;QACrB,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,OAAO,CAAC;KAC9B,CAAC;AACJ,CAAC"}

package/dist/auditor.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"auditor.d.ts","sourceRoot":"","sources":["../src/auditor.ts"],"names":[],"mappings":"AAoDA,OAAO,KAAK,EAAE,YAAY,EAAE,YAAY,EAAwG,MAAM,YAAY,CAAC;AA8wBnK,wBAAsB,WAAW,CAAC,MAAM,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,YAAY,GAAG,OAAO,CAAC,YAAY,CAAC,CAod/F"}
+{"version":3,"file":"auditor.d.ts","sourceRoot":"","sources":["../src/auditor.ts"],"names":[],"mappings":"AA6DA,OAAO,KAAK,EAAE,YAAY,EAAE,YAAY,EAAwG,MAAM,YAAY,CAAC;AAu0BnK,wBAAsB,WAAW,CAAC,MAAM,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,YAAY,GAAG,OAAO,CAAC,YAAY,CAAC,CAgf/F"}

package/dist/auditor.js

@@ -27,6 +27,14 @@ import { ogCompletenessRule } from "./rules/tech/og-completeness.js";
 import { robotsNoindexConflictRule } from "./rules/tech/robots-noindex-conflict.js";
 import { sitemapCompletenessRule } from "./rules/tech/sitemap-completeness.js";
 import { robotsComplianceRule, parseDisallowPatterns, isBlockedByPattern, parseCrawlDelaySeconds } from "./rules/tech/robots-sitemap-presence.js";
+import { llmsTxtRule } from "./rules/aeo/llms-txt.js";
+import { crawlerAccessRule } from "./rules/aeo/crawler-access.js";
+import { freshnessSignalsRule } from "./rules/aeo/freshness-signals.js";
+import { faqCoverageRule } from "./rules/aeo/faq-coverage.js";
+import { answerFirstRule } from "./rules/aeo/answer-first.js";
+import { citableFactsRule } from "./rules/aeo/citable-facts.js";
+import { nonReplicableValueRule } from "./rules/aeo/non-replicable-value.js";
+import { contentModularityRule } from "./rules/aeo/content-modularity.js";
 import { redirectChainRule } from "./rules/tech/redirect-chain.js";
 import { soft404Rule } from "./rules/tech/soft-404.js";
 import { jsonLdValidRule } from "./rules/schema/json-ld-valid.js";
@@ -38,6 +46,7 @@ import { urlPatternRule } from "./rules/cannibal/url-pattern.js";
 import { templateCoverageRule } from "./rules/spam/template-coverage.js";
 import { dataBindingRule, dataIdenticalRule } from "./rules/data/data-binding.js";
 import { classifyPages, isRuleEnabled } from "./page-classifier.js";
+import { isRuleAllowedInDiff } from "./rules/scope.js";
 import { RULE_REFERENCES } from "./rule-references.js";
 import { enrichFindings } from "./enrich-findings.js";
 import { triageFindings } from "./ai/triage.js";
@@ -61,13 +70,21 @@ const DEFAULTS = {
     hubPagesMaxSiblings: 50,
     titleOverlapThreshold: 0.8,
     keywordCollisionMinShared: 6,
-    templateCoverageMinPages: 5
+    templateCoverageMinPages: 5,
+    answerFirstMaxWords: 100,
+    citableFactsMin: 3,
+    citableFactsTarget: 8,
+    freshnessMaxStaleDays: 180,
+    modularityMaxParagraphWords: 200,
+    modularityMinSelfContainedRatio: 0.7,
+    faqMinQuestionHeadings: 2
 };
 const CATEGORY_WEIGHTS = {
-    spam: 0.4,
-    content: 0.25,
-    links: 0.15,
-    tech: 0.1,
+    spam: 0.35,
+    content: 0.2,
+    aeo: 0.15,
+    links: 0.12,
+    tech: 0.08,
     schema: 0.05,
     cannibal: 0.05,
     /** Dedup / crawl hygiene; does not affect composite score. */
@@ -93,8 +110,9 @@ function resolveGroupRules(baseRules, overrides) {
     }
     return result;
 }
-function runRulesOnPages(pages, resolvedRules, isEnabled, groupName, knownUrls, adjacency, inbound, rootUrl, normalizeUrlOptions, source, entityPatterns, overrides) {
+function runRulesOnPages(pages, resolvedRules, isEnabled, groupName, knownUrls, adjacency, inbound, rootUrl, normalizeUrlOptions, source, entityPatterns, overrides, mode = "full") {
     const findings = [];
+    const modeOk = (ruleId) => mode !== "diff" || isRuleAllowedInDiff(ruleId);
     const tag = (results) => results.map((r) => {
         const override = overrides?.[r.ruleId];
         return {
@@ -106,106 +124,137 @@ function runRulesOnPages(pages, resolvedRules, isEnabled, groupName, knownUrls,
     });
     // Spam rules — always compute cross-page data, only push findings if enabled
     const nearDuplicate = nearDuplicateRule(pages, resolvedRules.nearDuplicateThreshold);
-    if (isEnabled("spam/near-duplicate")) {
+    if (isEnabled("spam/near-duplicate") && modeOk("spam/near-duplicate")) {
         findings.push(...tag(nearDuplicate.findings));
     }
     const entitySwap = entitySwapRule(pages, entityPatterns, resolvedRules.entitySwapThreshold);
-    if (isEnabled("spam/entity-swap")) {
+    if (isEnabled("spam/entity-swap") && modeOk("spam/entity-swap")) {
         findings.push(...tag(entitySwap.findings));
     }
     const thinContent = thinContentRule(pages, resolvedRules.thinContentMinWords);
-    if (isEnabled("spam/thin-content")) {
+    if (isEnabled("spam/thin-content") && modeOk("spam/thin-content")) {
         findings.push(...tag(thinContent.findings));
     }
-    if (isEnabled("spam/doorway-pattern")) {
+    if (isEnabled("spam/doorway-pattern") && modeOk("spam/doorway-pattern")) {
         findings.push(...tag(doorwayPatternRule(nearDuplicate.pairs, entitySwap.pairs, thinContent.thinContentUrls, pages)));
     }
-    if (isEnabled("spam/publication-velocity")) {
+    if (isEnabled("spam/publication-velocity") && modeOk("spam/publication-velocity")) {
         findings.push(...tag(publicationVelocityRule(pages, resolvedRules.publicationVelocityMaxPerDay)));
     }
-    if (isEnabled("spam/boilerplate-ratio")) {
+    if (isEnabled("spam/boilerplate-ratio") && modeOk("spam/boilerplate-ratio")) {
         findings.push(...tag(boilerplateRatioRule(pages, resolvedRules.boilerplateMaxRatio)));
     }
-    if (isEnabled("spam/template-diversity")) {
+    if (isEnabled("spam/template-diversity") && modeOk("spam/template-diversity")) {
         findings.push(...tag(templateDiversityRule(pages, resolvedRules.templateDiversityMinUniqueRatio)));
     }
-    if (isEnabled("spam/template-coverage")) {
+    if (isEnabled("spam/template-coverage") && modeOk("spam/template-coverage")) {
         findings.push(...tag(templateCoverageRule(pages, entityPatterns, resolvedRules.templateCoverageMinPages)));
     }
     // Content rules
-    if (isEnabled("content/unique-value")) {
+    if (isEnabled("content/unique-value") && modeOk("content/unique-value")) {
         findings.push(...tag(uniqueValueRule(pages, resolvedRules.uniqueValueMinWords)));
     }
-    if (isEnabled("content/heading-uniqueness")) {
+    if (isEnabled("content/heading-uniqueness") && modeOk("content/heading-uniqueness")) {
         findings.push(...tag(headingUniquenessRule(pages, entityPatterns)));
     }
-    if (isEnabled("content/meta-uniqueness")) {
+    if (isEnabled("content/meta-uniqueness") && modeOk("content/meta-uniqueness")) {
         findings.push(...tag(metaUniquenessRule(pages, entityPatterns, resolvedRules.metaUniquenessMinJaccard)));
     }
-    if (isEnabled("content/missing-author")) {
+    if (isEnabled("content/missing-author") && modeOk("content/missing-author")) {
         findings.push(...tag(missingAuthorRule(pages)));
     }
-    if (isEnabled("content/eeat-signals")) {
+    if (isEnabled("content/eeat-signals") && modeOk("content/eeat-signals")) {
         findings.push(...tag(eeatSignalsRule(pages)));
     }
     // Link rules — use the global link graph
-    if (isEnabled("links/orphan-pages")) {
+    if (isEnabled("links/orphan-pages") && modeOk("links/orphan-pages")) {
         findings.push(...tag(orphanPagesRule(pages, inbound, rootUrl)));
     }
-    if (isEnabled("links/dead-ends")) {
+    if (isEnabled("links/dead-ends") && modeOk("links/dead-ends")) {
         findings.push(...tag(deadEndsRule(pages, knownUrls, rootUrl)));
     }
-    if (isEnabled("links/link-depth")) {
+    if (isEnabled("links/link-depth") && modeOk("links/link-depth")) {
         if (rootUrl) {
             findings.push(...tag(linkDepthRule(pages, adjacency, rootUrl, resolvedRules.linkDepthMaxClicks, inbound)));
         }
     }
-    if (isEnabled("links/cluster-connectivity")) {
+    if (isEnabled("links/cluster-connectivity") && modeOk("links/cluster-connectivity")) {
         findings.push(...tag(clusterConnectivityRule(pages, knownUrls)));
     }
-    if (isEnabled("links/hub-pages")) {
+    if (isEnabled("links/hub-pages") && modeOk("links/hub-pages")) {
         findings.push(...tag(hubPagesRule(pages, knownUrls, resolvedRules.hubPagesMinSiblings, resolvedRules.hubPagesMaxSiblings)));
     }
     // Tech rules
-    if (isEnabled("tech/canonical-consistency")) {
+    if (isEnabled("tech/canonical-consistency") && modeOk("tech/canonical-consistency")) {
         findings.push(...tag(canonicalConsistencyRule(pages, knownUrls, normalizeUrlOptions)));
     }
-    if (isEnabled("tech/canonical-noindex-conflict")) {
+    if (isEnabled("tech/canonical-noindex-conflict") && modeOk("tech/canonical-noindex-conflict")) {
         findings.push(...tag(canonicalNoindexConflictRule(pages, normalizeUrlOptions)));
     }
-    if (isEnabled("tech/robots-noindex-conflict")) {
+    if (isEnabled("tech/robots-noindex-conflict") && modeOk("tech/robots-noindex-conflict")) {
         findings.push(...tag(robotsNoindexConflictRule(pages, inbound)));
     }
-    if (isEnabled("tech/redirect-chain")) {
+    if (isEnabled("tech/redirect-chain") && modeOk("tech/redirect-chain")) {
         findings.push(...tag(redirectChainRule(pages)));
     }
-    if (isEnabled("tech/soft-404")) {
+    if (isEnabled("tech/soft-404") && modeOk("tech/soft-404")) {
         findings.push(...tag(soft404Rule(pages)));
     }
-    if (isEnabled("tech/og-completeness")) {
+    if (isEnabled("tech/og-completeness") && modeOk("tech/og-completeness")) {
         findings.push(...tag(ogCompletenessRule(pages)));
     }
-    if (isEnabled("tech/hreflang-consistency")) {
+    if (isEnabled("tech/hreflang-consistency") && modeOk("tech/hreflang-consistency")) {
         findings.push(...tag(hreflangConsistencyRule(pages, normalizeUrlOptions)));
     }
     // Schema rules
-    if (isEnabled("schema/json-ld-valid")) {
+    if (isEnabled("schema/json-ld-valid") && modeOk("schema/json-ld-valid")) {
         findings.push(...tag(jsonLdValidRule(pages)));
     }
-    if (isEnabled("schema/required-fields")) {
+    if (isEnabled("schema/required-fields") && modeOk("schema/required-fields")) {
         findings.push(...tag(requiredFieldsRule(pages)));
     }
-    if (isEnabled("schema/consistency")) {
+    if (isEnabled("schema/consistency") && modeOk("schema/consistency")) {
         findings.push(...tag(schemaConsistencyRule(pages)));
     }
+    // AEO rules
+    if (isEnabled("aeo/freshness-signals")) {
+        findings.push(...tag(freshnessSignalsRule(pages, {
+            maxStaleDays: resolvedRules.freshnessMaxStaleDays,
+        })));
+    }
+    if (isEnabled("aeo/faq-coverage")) {
+        findings.push(...tag(faqCoverageRule(pages, {
+            minQuestionHeadings: resolvedRules.faqMinQuestionHeadings,
+        })));
+    }
+    if (isEnabled("aeo/answer-first")) {
+        findings.push(...tag(answerFirstRule(pages, entityPatterns, {
+            maxFirstParagraphWords: resolvedRules.answerFirstMaxWords,
+        })));
+    }
+    if (isEnabled("aeo/citable-facts")) {
+        findings.push(...tag(citableFactsRule(pages, entityPatterns, {
+            minFactsPerPage: resolvedRules.citableFactsMin,
+            targetFactsPerPage: resolvedRules.citableFactsTarget,
+        })));
+    }
+    if (isEnabled("aeo/non-replicable-value")) {
+        findings.push(...tag(nonReplicableValueRule(pages)));
+    }
+    if (isEnabled("aeo/content-modularity")) {
+        findings.push(...tag(contentModularityRule(pages, {
+            maxParagraphWords: resolvedRules.modularityMaxParagraphWords,
+            minSelfContainedRatio: resolvedRules.modularityMinSelfContainedRatio,
+        })));
+    }
     // Cannibal rules
-    if (isEnabled("cannibal/title-overlap")) {
+    if (isEnabled("cannibal/title-overlap") && modeOk("cannibal/title-overlap")) {
         findings.push(...tag(titleOverlapRule(pages, entityPatterns, resolvedRules.titleOverlapThreshold)));
     }
-    if (isEnabled("cannibal/keyword-collision")) {
+    if (isEnabled("cannibal/keyword-collision") && modeOk("cannibal/keyword-collision")) {
         findings.push(...tag(keywordCollisionRule(pages, resolvedRules.keywordCollisionMinShared)));
     }
-    if (isEnabled("cannibal/url-pattern")) {
+    if (isEnabled("cannibal/url-pattern") && modeOk("cannibal/url-pattern")) {
         findings.push(...tag(urlPatternRule(pages)));
     }
     return findings;
@@ -223,6 +272,7 @@ function scoreFromFindings(findings) {
     const raw = {
         spam: 0,
         content: 0,
+        aeo: 0,
         links: 0,
         tech: 0,
         schema: 0,
@@ -238,6 +288,7 @@ function scoreFromFindings(findings) {
     }
     const weighted = raw.spam * CATEGORY_WEIGHTS.spam +
         raw.content * CATEGORY_WEIGHTS.content +
+        raw.aeo * CATEGORY_WEIGHTS.aeo +
         raw.links * CATEGORY_WEIGHTS.links +
         raw.tech * CATEGORY_WEIGHTS.tech +
         raw.schema * CATEGORY_WEIGHTS.schema +
@@ -248,6 +299,7 @@ function scoreFromFindings(findings) {
         categoryScores: {
             spam: raw.spam,
             content: raw.content,
+            aeo: raw.aeo,
             links: raw.links,
             tech: raw.tech,
             schema: raw.schema,
@@ -702,7 +754,14 @@ export async function auditSource(source, options) {
         hubPagesMaxSiblings: options?.rules?.hubPagesMaxSiblings ?? DEFAULTS.hubPagesMaxSiblings,
         titleOverlapThreshold: options?.rules?.titleOverlapThreshold ?? DEFAULTS.titleOverlapThreshold,
         keywordCollisionMinShared: options?.rules?.keywordCollisionMinShared ?? DEFAULTS.keywordCollisionMinShared,
-        templateCoverageMinPages: options?.rules?.templateCoverageMinPages ?? DEFAULTS.templateCoverageMinPages
+        templateCoverageMinPages: options?.rules?.templateCoverageMinPages ?? DEFAULTS.templateCoverageMinPages,
+        answerFirstMaxWords: options?.rules?.answerFirstMaxWords ?? DEFAULTS.answerFirstMaxWords,
+        citableFactsMin: options?.rules?.citableFactsMin ?? DEFAULTS.citableFactsMin,
+        citableFactsTarget: options?.rules?.citableFactsTarget ?? DEFAULTS.citableFactsTarget,
+        freshnessMaxStaleDays: options?.rules?.freshnessMaxStaleDays ?? DEFAULTS.freshnessMaxStaleDays,
+        modularityMaxParagraphWords: options?.rules?.modularityMaxParagraphWords ?? DEFAULTS.modularityMaxParagraphWords,
+        modularityMinSelfContainedRatio: options?.rules?.modularityMinSelfContainedRatio ?? DEFAULTS.modularityMinSelfContainedRatio,
+        faqMinQuestionHeadings: options?.rules?.faqMinQuestionHeadings ?? DEFAULTS.faqMinQuestionHeadings
     };
     const normalizeUrlOptions = mergeNormalizeUrlOptions({
         stripQuery: options?.rules?.stripUrlQuery ?? true,
@@ -836,9 +895,14 @@ export async function auditSource(source, options) {
                     throw new Error(`Invalid regex flags "${rawFlags}" in entityPatterns for placeholder "${p.placeholder}". ` +
                         `Only the flags g, i, m, s, u, y are permitted.`);
                 }
+                // Entity patterns are used with String.replace to mask every occurrence, which
+                // requires the `g` flag. Add it if the user forgot — a silently broken "only first
+                // match masked" regex would make template-detection rules (answer-first,
+                // citable-facts) miss shared openers.
+                const normalizedFlags = rawFlags.includes("g") ? rawFlags : `${rawFlags}g`;
                 try {
                     // Flags validated against SAFE_FLAGS_RE above; pattern is from trusted local config, not HTTP input.
-                    return { placeholder: p.placeholder, pattern: new RegExp(p.pattern, rawFlags) }; // nosemgrep
+                    return { placeholder: p.placeholder, pattern: new RegExp(p.pattern, normalizedFlags) }; // nosemgrep
                 }
                 catch (err) {
                     throw new Error(`Invalid regex pattern for placeholder "${p.placeholder}": ${err.message}`);
@@ -851,8 +915,9 @@ export async function auditSource(source, options) {
     const allFindings = [...duplicateUrlFindings];
     const groupScores = {};
     const groupPageCounts = {};
+    const auditMode = options?.mode ?? "full";
     // Site-wide rules (run once, outside group loop)
-    if (sitemapUrlSet && sitemapUrlSet.size > 0) {
+    if (sitemapUrlSet && sitemapUrlSet.size > 0 && auditMode !== "diff") {
         const sitemapFindings = sitemapCompletenessRule(parsedPages, sitemapUrlSet);
         allFindings.push(...sitemapFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
         if (robotsTxtContent) {
@@ -860,13 +925,24 @@ export async function auditSource(source, options) {
             allFindings.push(...robotsFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
         }
     }
+    // AEO site-wide rules. These run unconditionally (consistent with sitemap-completeness
+    // and robots-compliance); page-group rule lists govern per-page AEO rules only.
+    const llmsFindings = await llmsTxtRule(source, { timeoutMs });
+    allFindings.push(...llmsFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
+    if (robotsTxtContent) {
+        const crawlerFindings = crawlerAccessRule(robotsTxtContent);
+        allFindings.push(...crawlerFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
+    }
     // Data source comparison rules
     if (options?.dataSource?.records && options.dataSource.records.length > 0) {
-        const dataFindings = [
-            ...dataBindingRule(parsedPages, options.dataSource.records),
-            ...dataIdenticalRule(parsedPages, options.dataSource.records),
-        ];
-        allFindings.push(...dataFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
+        if (auditMode !== "diff" || isRuleAllowedInDiff("data/missing-binding")) {
+            const dataBindingFindings = dataBindingRule(parsedPages, options.dataSource.records);
+            allFindings.push(...dataBindingFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
+        }
+        if (auditMode !== "diff" || isRuleAllowedInDiff("data/identical-across-pages")) {
+            const dataIdenticalFindings = dataIdenticalRule(parsedPages, options.dataSource.records);
+            allFindings.push(...dataIdenticalFindings.map((f) => ({ ...f, ref: f.ref ?? RULE_REFERENCES[f.ruleId] })));
+        }
     }
     for (const [groupName, groupPages] of classified) {
         if (groupPages.length === 0)
@@ -876,7 +952,7 @@ export async function auditSource(source, options) {
             continue;
         const groupRules = resolveGroupRules(resolvedRules, groupConfig?.overrides);
         const enabledCheck = (ruleId) => isRuleEnabled(ruleId, groupConfig?.rules);
-        const findings = runRulesOnPages(groupPages, groupRules, enabledCheck, groupName, knownUrls, adjacency, inbound, rootUrl, normalizeUrlOptions, source, DEFAULT_ENTITY_PATTERNS, groupConfig?.overrides);
+        const findings = runRulesOnPages(groupPages, groupRules, enabledCheck, groupName, knownUrls, adjacency, inbound, rootUrl, normalizeUrlOptions, source, DEFAULT_ENTITY_PATTERNS, groupConfig?.overrides, options?.mode ?? "full");
         allFindings.push(...findings);
         groupPageCounts[groupName] = groupPages.length;
         const { score } = scoreFromFindings(findings);