npm - @rankcli/agent-runtime - Versions diffs - 0.0.15 → 0.0.16 - Mend

@rankcli/agent-runtime 0.0.15 → 0.0.16

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (202) hide show

package/dist/index.d.mts +43 -393
package/dist/index.d.ts +43 -393
package/dist/index.js +306 -6526
package/dist/index.mjs +202 -6412
package/package.json +6 -1
package/scripts/build-deno.ts +0 -134
package/src/analyzers/core-web-vitals-analyzer.test.ts +0 -236
package/src/analyzers/core-web-vitals-analyzer.ts +0 -557
package/src/analyzers/geo-analyzer.test.ts +0 -310
package/src/analyzers/geo-analyzer.ts +0 -814
package/src/analyzers/image-optimization-analyzer.test.ts +0 -145
package/src/analyzers/image-optimization-analyzer.ts +0 -348
package/src/analyzers/index.ts +0 -233
package/src/analyzers/internal-linking-analyzer.test.ts +0 -141
package/src/analyzers/internal-linking-analyzer.ts +0 -419
package/src/analyzers/mobile-seo-analyzer.test.ts +0 -140
package/src/analyzers/mobile-seo-analyzer.ts +0 -455
package/src/analyzers/security-headers-analyzer.test.ts +0 -115
package/src/analyzers/security-headers-analyzer.ts +0 -318
package/src/analyzers/structured-data-analyzer.test.ts +0 -210
package/src/analyzers/structured-data-analyzer.ts +0 -590
package/src/audit/ai/analyzer.ts +0 -347
package/src/audit/ai/index.ts +0 -29
package/src/audit/ai/prompts/content-analysis.ts +0 -271
package/src/audit/ai/types.ts +0 -179
package/src/audit/checks/additional-checks.ts +0 -439
package/src/audit/checks/agent-experience.ts +0 -108
package/src/audit/checks/ai-citation-worthiness.ts +0 -399
package/src/audit/checks/ai-content-structure.ts +0 -325
package/src/audit/checks/ai-readiness.ts +0 -406
package/src/audit/checks/anchor-text.ts +0 -179
package/src/audit/checks/answer-conciseness.ts +0 -322
package/src/audit/checks/asset-minification.ts +0 -270
package/src/audit/checks/bing-optimization.ts +0 -206
package/src/audit/checks/brand-mention-optimization.ts +0 -349
package/src/audit/checks/caching-headers.ts +0 -305
package/src/audit/checks/canonical-advanced.ts +0 -150
package/src/audit/checks/canonical-domain.ts +0 -196
package/src/audit/checks/citation-quality.ts +0 -358
package/src/audit/checks/client-rendering.ts +0 -542
package/src/audit/checks/color-contrast.ts +0 -342
package/src/audit/checks/content-freshness.ts +0 -170
package/src/audit/checks/content-science.ts +0 -589
package/src/audit/checks/conversion-elements.ts +0 -526
package/src/audit/checks/crawlability.ts +0 -220
package/src/audit/checks/directory-listing.ts +0 -172
package/src/audit/checks/dom-analysis.ts +0 -191
package/src/audit/checks/dom-size.ts +0 -246
package/src/audit/checks/duplicate-content.ts +0 -194
package/src/audit/checks/eeat-signals.ts +0 -990
package/src/audit/checks/entity-seo.ts +0 -396
package/src/audit/checks/featured-snippet.ts +0 -473
package/src/audit/checks/freshness-signals.ts +0 -443
package/src/audit/checks/funnel-intent.ts +0 -463
package/src/audit/checks/hreflang.ts +0 -174
package/src/audit/checks/html-compliance.ts +0 -302
package/src/audit/checks/image-dimensions.ts +0 -167
package/src/audit/checks/images.ts +0 -160
package/src/audit/checks/indexnow.ts +0 -275
package/src/audit/checks/interactive-tools.ts +0 -475
package/src/audit/checks/internal-link-graph.ts +0 -436
package/src/audit/checks/keyword-analysis.ts +0 -239
package/src/audit/checks/keyword-cannibalization.ts +0 -385
package/src/audit/checks/keyword-placement.ts +0 -471
package/src/audit/checks/links.ts +0 -273
package/src/audit/checks/llms-txt.ts +0 -224
package/src/audit/checks/local-seo.ts +0 -296
package/src/audit/checks/mobile.ts +0 -167
package/src/audit/checks/modern-images.ts +0 -226
package/src/audit/checks/navboost-signals.ts +0 -395
package/src/audit/checks/on-page.ts +0 -209
package/src/audit/checks/page-resources.ts +0 -285
package/src/audit/checks/pagination.ts +0 -180
package/src/audit/checks/performance.ts +0 -153
package/src/audit/checks/platform-presence.ts +0 -580
package/src/audit/checks/rag-chunk-readiness.test.ts +0 -159
package/src/audit/checks/rag-chunk-readiness.ts +0 -163
package/src/audit/checks/redirect-analysis.ts +0 -153
package/src/audit/checks/redirect-chain.ts +0 -389
package/src/audit/checks/resource-hints.ts +0 -420
package/src/audit/checks/responsive-css.ts +0 -247
package/src/audit/checks/responsive-images.ts +0 -396
package/src/audit/checks/review-ecosystem.ts +0 -415
package/src/audit/checks/robots-validation.ts +0 -373
package/src/audit/checks/security-headers.ts +0 -188
package/src/audit/checks/security.ts +0 -144
package/src/audit/checks/serp-preview.ts +0 -251
package/src/audit/checks/site-maturity.ts +0 -453
package/src/audit/checks/social-meta.test.ts +0 -275
package/src/audit/checks/social-meta.ts +0 -134
package/src/audit/checks/soft-404.ts +0 -151
package/src/audit/checks/structured-data.ts +0 -238
package/src/audit/checks/tech-detection.ts +0 -496
package/src/audit/checks/topical-clusters.ts +0 -435
package/src/audit/checks/tracker-bloat.ts +0 -462
package/src/audit/checks/tracking-verification.test.ts +0 -371
package/src/audit/checks/tracking-verification.ts +0 -636
package/src/audit/checks/url-safety.ts +0 -682
package/src/audit/deno-entry.ts +0 -66
package/src/audit/discovery/index.ts +0 -15
package/src/audit/discovery/link-crawler.ts +0 -233
package/src/audit/discovery/repo-routes.ts +0 -347
package/src/audit/engine.ts +0 -633
package/src/audit/fixes/index.ts +0 -209
package/src/audit/fixes/social-meta-fixes.test.ts +0 -329
package/src/audit/fixes/social-meta-fixes.ts +0 -463
package/src/audit/index.ts +0 -74
package/src/audit/runner.test.ts +0 -299
package/src/audit/runner.ts +0 -130
package/src/audit/types.ts +0 -1987
package/src/content/featured-snippet.ts +0 -367
package/src/content/generator.test.ts +0 -534
package/src/content/generator.ts +0 -501
package/src/content/headline.ts +0 -317
package/src/content/index.ts +0 -62
package/src/content/intent.ts +0 -258
package/src/content/keyword-density.ts +0 -349
package/src/content/readability.ts +0 -262
package/src/executor.ts +0 -336
package/src/fixer/framework-fixes.test.ts +0 -489
package/src/fixer/framework-fixes.ts +0 -5287
package/src/fixer/index.ts +0 -6
package/src/fixer/schemas.ts +0 -971
package/src/fixer.ts +0 -642
package/src/frameworks/detector.test.ts +0 -248
package/src/frameworks/detector.ts +0 -899
package/src/frameworks/index.ts +0 -68
package/src/frameworks/recipes/angular.yaml +0 -171
package/src/frameworks/recipes/astro.yaml +0 -206
package/src/frameworks/recipes/django.yaml +0 -180
package/src/frameworks/recipes/laravel.yaml +0 -137
package/src/frameworks/recipes/nextjs.yaml +0 -268
package/src/frameworks/recipes/nuxt.yaml +0 -175
package/src/frameworks/recipes/rails.yaml +0 -188
package/src/frameworks/recipes/react.yaml +0 -202
package/src/frameworks/recipes/sveltekit.yaml +0 -154
package/src/frameworks/recipes/vue.yaml +0 -137
package/src/frameworks/recipes/wordpress.yaml +0 -209
package/src/frameworks/suggestion-engine.ts +0 -357
package/src/geo/geo-content.test.ts +0 -305
package/src/geo/geo-content.ts +0 -266
package/src/geo/geo-history.test.ts +0 -473
package/src/geo/geo-history.ts +0 -433
package/src/geo/geo-tracker.test.ts +0 -359
package/src/geo/geo-tracker.ts +0 -411
package/src/geo/index.ts +0 -11
package/src/geo/llm-citation-checker.ts +0 -188
package/src/git/commit-helper.test.ts +0 -261
package/src/git/commit-helper.ts +0 -329
package/src/git/index.ts +0 -12
package/src/git/pr-helper.test.ts +0 -284
package/src/git/pr-helper.ts +0 -307
package/src/index.ts +0 -76
package/src/keywords/ai-keyword-engine.ts +0 -1062
package/src/keywords/ai-summarizer.ts +0 -387
package/src/keywords/ci-mode.ts +0 -555
package/src/keywords/engine.ts +0 -359
package/src/keywords/index.ts +0 -151
package/src/keywords/llm-judge.ts +0 -357
package/src/keywords/nlp-analysis.ts +0 -706
package/src/keywords/prioritizer.ts +0 -295
package/src/keywords/site-crawler.ts +0 -342
package/src/keywords/sources/autocomplete.ts +0 -139
package/src/keywords/sources/competitive-search.ts +0 -450
package/src/keywords/sources/competitor-analysis.ts +0 -374
package/src/keywords/sources/dataforseo.ts +0 -206
package/src/keywords/sources/free-sources.ts +0 -294
package/src/keywords/sources/gsc.ts +0 -123
package/src/keywords/topic-grouping.ts +0 -327
package/src/keywords/types.ts +0 -144
package/src/keywords/wizard.ts +0 -457
package/src/loader.ts +0 -40
package/src/ranking/index.ts +0 -5
package/src/ranking/serp-client.ts +0 -348
package/src/ranking/tracker.ts +0 -380
package/src/ranking/types.ts +0 -123
package/src/reports/index.ts +0 -7
package/src/reports/report-generator.test.ts +0 -293
package/src/reports/report-generator.ts +0 -713
package/src/scheduler/alerts.test.ts +0 -458
package/src/scheduler/alerts.ts +0 -328
package/src/scheduler/index.ts +0 -8
package/src/scheduler/scheduled-audit.test.ts +0 -377
package/src/scheduler/scheduled-audit.ts +0 -149
package/src/test/integration-test.ts +0 -325
package/src/tools/analyzer.ts +0 -373
package/src/tools/crawl.ts +0 -293
package/src/tools/files.ts +0 -301
package/src/tools/h1-fixer.ts +0 -249
package/src/tools/index.ts +0 -67
package/src/tracking/github-action.ts +0 -326
package/src/tracking/google-analytics.ts +0 -265
package/src/tracking/index.ts +0 -45
package/src/tracking/report-generator.ts +0 -386
package/src/tracking/search-console.ts +0 -335
package/src/types.ts +0 -148
package/src/utils/http.ts +0 -302
package/src/wasm-adapter.ts +0 -297
package/src/wasm-entry.ts +0 -14
package/tsconfig.json +0 -17
package/tsup.wasm.config.ts +0 -26
package/vitest.config.ts +0 -15

package/src/audit/checks/rag-chunk-readiness.test.ts DELETED Viewed

@@ -1,159 +0,0 @@
-import { describe, it, expect } from 'vitest';
-import { analyzeRAGChunkReadiness } from './rag-chunk-readiness.js';
-const testUrl = 'https://example.com';
-function words(n: number, prefix = 'word'): string {
-  return Array.from({ length: n }, (_, i) => `${prefix}${i}`).join(' ');
-}
-describe('analyzeRAGChunkReadiness', () => {
-  it('returns zero sections and score 0 for a page with no headings', () => {
-    const html = `<html><body><p>${words(200)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.totalSections).toBe(0);
-    expect(data.chunkReadinessScore).toBe(0);
-  });
-  it('classifies a well-sized section (150-450 words) as good', () => {
-    const html = `<html><body><h2>Topic</h2><p>${words(300)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.totalSections).toBe(1);
-    expect(data.sections[0].wordCount).toBe(300);
-    expect(data.sections[0].sizeQuality).toBe('good');
-  });
-  it('classifies a very short section as too-short', () => {
-    const html = `<html><body><h2>Topic</h2><p>${words(20)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.sections[0].sizeQuality).toBe('too-short');
-  });
-  it('classifies a very long section as too-long', () => {
-    const html = `<html><body><h2>Topic</h2><p>${words(800)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.sections[0].sizeQuality).toBe('too-long');
-  });
-  it('splits content into sections at each h1/h2/h3, regardless of nesting depth', () => {
-    // Deliberately nested in wrapper divs — a sibling-only DOM walk would
-    // fail to associate the paragraphs with their heading here; document
-    // order should still get this right.
-    const html = `
-      <html><body>
-        <div class="section-wrapper">
-          <div class="inner"><h2>First</h2></div>
-          <div class="content"><p>${words(200, 'a')}</p></div>
-        </div>
-        <div class="section-wrapper">
-          <div class="inner"><h2>Second</h2></div>
-          <div class="content"><p>${words(200, 'b')}</p></div>
-        </div>
-      </body></html>
-    `;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.totalSections).toBe(2);
-    expect(data.sections[0].heading).toBe('First');
-    expect(data.sections[0].wordCount).toBe(200);
-    expect(data.sections[1].heading).toBe('Second');
-    expect(data.sections[1].wordCount).toBe(200);
-  });
-  it('accumulates multiple paragraphs and list items under one heading', () => {
-    const html = `
-      <html><body>
-        <h2>Topic</h2>
-        <p>${words(100, 'a')}</p>
-        <p>${words(100, 'b')}</p>
-        <ul><li>${words(50, 'c')}</li><li>${words(50, 'd')}</li></ul>
-      </body></html>
-    `;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.totalSections).toBe(1);
-    expect(data.sections[0].wordCount).toBe(300);
-  });
-  it('content before the first heading is not attributed to any section', () => {
-    const html = `
-      <html><body>
-        <p>${words(500)}</p>
-        <h2>Topic</h2>
-        <p>${words(200)}</p>
-      </body></html>
-    `;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.totalSections).toBe(1);
-    expect(data.sections[0].wordCount).toBe(200);
-  });
-  it('detects a dangling-reference opener', () => {
-    const html = `<html><body><h2>Topic</h2><p>This approach reduces cost significantly. ${words(150)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.sections[0].startsWithDanglingReference).toBe(true);
-  });
-  it('does not flag a section that names its subject', () => {
-    const html = `<html><body><h2>Topic</h2><p>Caching reduces cost significantly. ${words(150)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.sections[0].startsWithDanglingReference).toBe(false);
-  });
-  it('raises RAG_CHUNK_SIZE_MISMATCH when most sections are poorly sized', () => {
-    const html = `
-      <html><body>
-        <h2>One</h2><p>${words(20)}</p>
-        <h2>Two</h2><p>${words(900)}</p>
-        <h2>Three</h2><p>${words(15)}</p>
-      </body></html>
-    `;
-    const { issues } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(issues.some((i) => i.code === 'RAG_CHUNK_SIZE_MISMATCH')).toBe(true);
-  });
-  it('does not raise RAG_CHUNK_SIZE_MISMATCH when most sections are well-sized', () => {
-    const html = `
-      <html><body>
-        <h2>One</h2><p>${words(200)}</p>
-        <h2>Two</h2><p>${words(250)}</p>
-        <h2>Three</h2><p>${words(300)}</p>
-      </body></html>
-    `;
-    const { issues } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(issues.some((i) => i.code === 'RAG_CHUNK_SIZE_MISMATCH')).toBe(false);
-  });
-  it('raises RAG_CHUNK_DANGLING_REFERENCES when over 30% of sections open with one', () => {
-    const html = `
-      <html><body>
-        <h2>One</h2><p>This reduces cost. ${words(150)}</p>
-        <h2>Two</h2><p>It also improves speed. ${words(150)}</p>
-        <h2>Three</h2><p>Caching is the mechanism. ${words(150)}</p>
-      </body></html>
-    `;
-    const { issues } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(issues.some((i) => i.code === 'RAG_CHUNK_DANGLING_REFERENCES')).toBe(true);
-  });
-  it('does not raise issues for a page with too few sections to judge (under 3)', () => {
-    const html = `<html><body><h2>One</h2><p>${words(20)}</p><h2>Two</h2><p>${words(900)}</p></body></html>`;
-    const { issues } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(issues.length).toBe(0);
-  });
-  it('computes a chunkReadinessScore of 100 when all sections are good and self-contained', () => {
-    const html = `
-      <html><body>
-        <h2>One</h2><p>Caching reduces latency. ${words(200)}</p>
-        <h2>Two</h2><p>Compression reduces bandwidth. ${words(200)}</p>
-      </body></html>
-    `;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.chunkReadinessScore).toBe(100);
-  });
-  it('estimates tokens at roughly 1.33 tokens per word', () => {
-    const html = `<html><body><h2>Topic</h2><p>${words(300)}</p></body></html>`;
-    const { data } = analyzeRAGChunkReadiness(html, testUrl);
-    expect(data.sections[0].estimatedTokens).toBe(399); // round(300 * 1.33)
-  });
-});

package/src/audit/checks/rag-chunk-readiness.ts DELETED Viewed

@@ -1,163 +0,0 @@
-/**
- * RAG Chunk Readiness
- *
- * How AI answer engines actually retrieve content: RAG (Retrieval-Augmented
- * Generation) systems chunk a page into pieces — the 2026 convergence
- * benchmark is roughly 400-512 tokens per chunk, 10-25% overlap — then
- * retrieve whichever chunk(s) best match a query, not the whole page. A page
- * whose own heading-delimited sections are naturally chunk-sized, with clear
- * headings and paragraphs that don't lean on the previous section's context
- * to make sense, is less likely to get split mid-thought by a RAG chunker —
- * preserving the semantic boundaries the author actually wrote, which is
- * what the RAG-chunking literature credits with better retrieval recall.
- *
- * This is a genuinely different angle from ai-content-structure.ts (tables/
- * lists/Q&A formatting): that's about FORMAT, this is about SECTION SIZE
- * and SELF-CONTAINMENT. Novel enough that no dedicated site-auditing tool
- * for it was found as of this writing — RAG-chunking guidance that exists
- * targets people building RAG systems, not people writing web content.
- */
-import * as cheerio from 'cheerio';
-import type { AuditIssue } from '../types.js';
-export interface ChunkSection {
-  heading: string;
-  headingLevel: number;
-  wordCount: number;
-  estimatedTokens: number;
-  sizeQuality: 'too-short' | 'good' | 'too-long';
-  startsWithDanglingReference: boolean;
-}
-export interface RAGChunkReadinessData {
-  sections: ChunkSection[];
-  totalSections: number;
-  wellSizedSectionCount: number;
-  danglingReferenceCount: number;
-  chunkReadinessScore: number; // 0-100
-}
-// 400-512 tokens is the 2026 RAG-chunking convergence benchmark. English
-// prose runs roughly 1.3 words per token (~0.75 tokens/word), so that's
-// roughly 300-385 words — widened to 150-450 here since this measures how
-// naturally an AUTHOR sectioned their content, not what a chunker itself
-// targets; a section in this band is unlikely to need mid-thought splitting
-// either way, which is the property that actually matters.
-const MIN_GOOD_WORDS = 150;
-const MAX_GOOD_WORDS = 450;
-const TOKENS_PER_WORD = 1.33;
-// A section opening on a bare pronoun/demonstrative reference ("This
-// approach...", "It also...") reads fine in place but loses its antecedent
-// the moment a RAG system retrieves it standalone — a real self-containment
-// signal, not a grammar nitpick.
-const DANGLING_OPENERS = /^(this|it|these|that|they|such|those|the former|the latter)\b/i;
-function classifySize(wordCount: number): ChunkSection['sizeQuality'] {
-  if (wordCount < MIN_GOOD_WORDS) return 'too-short';
-  if (wordCount > MAX_GOOD_WORDS) return 'too-long';
-  return 'good';
-}
-/**
- * Splits content into heading-delimited sections by walking all
- * heading/content elements in DOCUMENT ORDER (a single multi-tag cheerio
- * selector), rather than via sibling traversal — sibling-only walks break
- * on the wrapper-div-heavy DOM nesting common in modern frameworks, where
- * a heading and "its" paragraphs aren't actually direct siblings. Document
- * order is robust to arbitrary nesting depth.
- */
-function extractSections($: cheerio.CheerioAPI): Array<{ heading: string; headingLevel: number; text: string }> {
-  const elements = $('h1, h2, h3, p, li, blockquote').toArray();
-  const sections: Array<{ heading: string; headingLevel: number; text: string }> = [];
-  let current: { heading: string; headingLevel: number; text: string } | null = null;
-  for (const el of elements) {
-    const tag = (el as { tagName?: string }).tagName?.toLowerCase();
-    if (tag === 'h1' || tag === 'h2' || tag === 'h3') {
-      if (current) sections.push(current);
-      current = { heading: $(el).text().trim(), headingLevel: parseInt(tag.slice(1), 10), text: '' };
-    } else if (current) {
-      current.text += ' ' + $(el).text();
-    }
-    // Content before the first heading has no section to attribute it to —
-    // intentionally skipped rather than inventing a synthetic "intro" chunk.
-  }
-  if (current) sections.push(current);
-  return sections;
-}
-export function analyzeRAGChunkReadiness(
-  html: string,
-  url: string
-): { issues: AuditIssue[]; data: RAGChunkReadinessData } {
-  const issues: AuditIssue[] = [];
-  const $ = cheerio.load(html);
-  $('nav, footer, aside, script, style, noscript, header').remove();
-  const rawSections = extractSections($);
-  const sections: ChunkSection[] = rawSections.map((s) => {
-    const words = s.text.trim().split(/\s+/).filter(Boolean);
-    const wordCount = words.length;
-    const estimatedTokens = Math.round(wordCount * TOKENS_PER_WORD);
-    const firstSentence = s.text.trim().split(/[.!?]/)[0] || '';
-    return {
-      heading: s.heading,
-      headingLevel: s.headingLevel,
-      wordCount,
-      estimatedTokens,
-      sizeQuality: classifySize(wordCount),
-      startsWithDanglingReference: DANGLING_OPENERS.test(firstSentence.trim()),
-    };
-  });
-  const totalSections = sections.length;
-  const wellSizedSectionCount = sections.filter((s) => s.sizeQuality === 'good').length;
-  const danglingReferenceCount = sections.filter((s) => s.startsWithDanglingReference).length;
-  const chunkReadinessScore =
-    totalSections === 0
-      ? 0
-      : Math.round(
-          ((wellSizedSectionCount / totalSections) * 0.7 +
-            ((totalSections - danglingReferenceCount) / totalSections) * 0.3) *
-            100
-        );
-  if (totalSections >= 3 && wellSizedSectionCount / totalSections < 0.5) {
-    const tooLong = sections.filter((s) => s.sizeQuality === 'too-long').length;
-    const tooShort = sections.filter((s) => s.sizeQuality === 'too-short').length;
-    issues.push({
-      code: 'RAG_CHUNK_SIZE_MISMATCH',
-      severity: 'notice',
-      category: 'ai-readiness',
-      title: 'Most sections are poorly sized for AI retrieval chunking',
-      description: `${wellSizedSectionCount} of ${totalSections} sections fall in a well-sized range for RAG retrieval (roughly 150-450 words); ${tooLong} are too long and likely to get split mid-thought, ${tooShort} are too short and likely to get merged with unrelated neighbors.`,
-      impact: 'When an AI answer engine retrieves a chunk of this page in response to a query, oversized sections risk being cut off mid-point and undersized ones risk losing standalone context — both reduce the odds the retrieved chunk reads coherently in an AI-generated answer.',
-      howToFix: 'Break up long sections (450+ words) with an additional H2/H3 subheading roughly every 300-400 words. Merge very short sections (under 150 words) into a neighboring section or expand them with enough context to stand alone.',
-      affectedUrls: [url],
-      details: { wellSizedSectionCount, totalSections, tooLong, tooShort },
-    });
-  }
-  if (totalSections >= 3 && danglingReferenceCount / totalSections > 0.3) {
-    issues.push({
-      code: 'RAG_CHUNK_DANGLING_REFERENCES',
-      severity: 'notice',
-      category: 'ai-readiness',
-      title: 'Several sections open with a reference to prior context',
-      description: `${danglingReferenceCount} of ${totalSections} sections start with a pronoun or demonstrative ("This...", "It...", "These...") that depends on the previous section to make sense.`,
-      impact: 'A RAG system that retrieves one of these sections on its own — which is exactly how retrieval works, one chunk at a time — surfaces a sentence whose subject is undefined, reading as broken or confusing in an AI-generated answer.',
-      howToFix: 'Open each section by naming its actual subject instead of referring back to the previous one — e.g. "This approach reduces cost" becomes "Caching reduces cost."',
-      affectedUrls: [url],
-      details: { danglingReferenceCount, totalSections },
-    });
-  }
-  return {
-    issues,
-    data: { sections, totalSections, wellSizedSectionCount, danglingReferenceCount, chunkReadinessScore },
-  };
-}

package/src/audit/checks/redirect-analysis.ts DELETED Viewed

@@ -1,153 +0,0 @@
-import { httpHead } from '../../utils/http.js';
-import type { AuditIssue } from '../types.js';
-import { ISSUE_DEFINITIONS } from '../types.js';
-export interface RedirectData {
-  hasRedirect: boolean;
-  redirectChain: RedirectHop[];
-  isLoop: boolean;
-  finalUrl?: string;
-  totalHops: number;
-}
-export interface RedirectHop {
-  url: string;
-  statusCode: number;
-  location?: string;
-}
-export async function analyzeRedirects(
-  url: string,
-  maxHops: number = 10
-): Promise<{ issues: AuditIssue[]; data: RedirectData }> {
-  const issues: AuditIssue[] = [];
-  const chain: RedirectHop[] = [];
-  const visitedUrls = new Set<string>();
-  let currentUrl = url;
-  let isLoop = false;
-  while (chain.length < maxHops) {
-    // Check for loop
-    if (visitedUrls.has(currentUrl)) {
-      isLoop = true;
-      break;
-    }
-    visitedUrls.add(currentUrl);
-    try {
-      const response = await httpHead(currentUrl, {
-        timeout: 5000,
-        maxRedirects: 0,
-        validateStatus: () => true,
-      });
-      const hop: RedirectHop = {
-        url: currentUrl,
-        statusCode: response.status,
-      };
-      if (response.status >= 300 && response.status < 400) {
-        const location = response.headers['location'];
-        if (location) {
-          hop.location = location;
-          chain.push(hop);
-          currentUrl = new URL(location, currentUrl).href;
-          continue;
-        }
-      }
-      chain.push(hop);
-      break;
-    } catch (error) {
-      chain.push({
-        url: currentUrl,
-        statusCode: 0,
-      });
-      break;
-    }
-  }
-  const hasRedirect = chain.length > 1;
-  const finalUrl = chain.length > 0 ? chain[chain.length - 1].url : url;
-  const data: RedirectData = {
-    hasRedirect,
-    redirectChain: chain,
-    isLoop,
-    finalUrl,
-    totalHops: chain.length - 1,
-  };
-  // ==================== Issue Detection ====================
-  // Redirect loop
-  if (isLoop) {
-    issues.push({
-      ...ISSUE_DEFINITIONS.REDIRECT_LOOP,
-      affectedUrls: [url],
-      details: {
-        chain: chain.map(h => `${h.url} (${h.statusCode})`),
-      },
-    });
-  }
-  // Redirect chain (more than one redirect)
-  if (!isLoop && chain.length > 2) {
-    issues.push({
-      ...ISSUE_DEFINITIONS.REDIRECT_CHAIN,
-      affectedUrls: [url],
-      details: {
-        chain: chain.map(h => `${h.url} (${h.statusCode})`),
-        hops: chain.length - 1,
-        finalUrl,
-      },
-    });
-  }
-  return { issues, data };
-}
-// Check internal links for redirect targets
-export async function checkInternalRedirects(
-  internalLinks: string[],
-  batchSize: number = 10
-): Promise<{ redirectLinks: Array<{ source: string; target: string; statusCode: number }> }> {
-  const redirectLinks: Array<{ source: string; target: string; statusCode: number }> = [];
-  // Process in batches to avoid overwhelming the server
-  for (let i = 0; i < internalLinks.length; i += batchSize) {
-    const batch = internalLinks.slice(i, i + batchSize);
-    const promises = batch.map(async (link) => {
-      try {
-        const response = await httpHead(link, {
-          timeout: 5000,
-          maxRedirects: 0,
-          validateStatus: () => true,
-        });
-        if (response.status >= 300 && response.status < 400) {
-          const location = response.headers['location'];
-          if (location) {
-            return {
-              source: link,
-              target: new URL(location, link).href,
-              statusCode: response.status,
-            };
-          }
-        }
-        return null;
-      } catch {
-        return null;
-      }
-    });
-    const results = await Promise.all(promises);
-    for (const result of results) {
-      if (result) {
-        redirectLinks.push(result);
-      }
-    }
-  }
-  return { redirectLinks };
-}