diff --git a/src/helpers/get-page-urls.ts b/src/helpers/get-page-urls.ts
index 90a86bc..2c2a137 100644
--- a/src/helpers/get-page-urls.ts
+++ b/src/helpers/get-page-urls.ts
@@ -238,9 +238,13 @@ async function discoverSitemapUrls(ctx: CheckContext, originOverride?: string):
   return candidates;
 }
 
+export type DiscoverySource = 'llms-txt' | 'sitemap' | 'fallback';
+
 export interface PageUrlResult {
   urls: string[];
   warnings: string[];
+  /** Which discovery methods contributed to the final URL set. */
+  sources: DiscoverySource[];
 }
 
 function isGzipped(url: string): boolean {
@@ -740,19 +744,40 @@ export function filterByPathPrefix(urls: string[], baseUrl: string): string[] {
   return urls.filter((url) => matchesPathPrefix(url, baseUrlPath));
 }
 
+/**
+ * Merge two URL arrays, preserving order. Primary URLs come first;
+ * secondary URLs are appended only if not already present.
+ */
+function mergeUrlSets(primary: string[], secondary: string[]): string[] {
+  const seen = new Set(primary);
+  const merged = [...primary];
+  for (const url of secondary) {
+    if (!seen.has(url)) {
+      merged.push(url);
+      seen.add(url);
+    }
+  }
+  return merged;
+}
+
 /**
  * Discover page URLs from llms.txt links, sitemap, or fall back to baseUrl.
  *
  * Priority:
- * 1. llms.txt links (from previous check results)
+ * 1. llms.txt links (from previous check results or direct fetch)
  * 2. Sitemap URLs (robots.txt Sitemap directives, then /sitemap.xml fallback)
  * 3. baseUrl fallback
  *
+ * When llms.txt produces URLs but fewer than `maxLinksToTest`, sitemap
+ * URLs are merged in (deduped) so the sample covers a broader surface.
+ * The `sources` field records which discovery methods contributed.
+ *
  * All discovered URLs are filtered to the baseUrl's path prefix so that
  * docs at a subpath (e.g. `/docs`) don't include unrelated site content.
  */
 export async function getPageUrls(ctx: CheckContext): Promise<PageUrlResult> {
   const warnings: string[] = [];
+  const sources: DiscoverySource[] = [];
 
   const locale = ctx.options.preferredLocale ?? extractLocaleFromUrl(ctx.baseUrl);
   const version = ctx.options.preferredVersion ?? extractVersionFromUrl(ctx.baseUrl);
@@ -770,22 +795,43 @@ export async function getPageUrls(ctx: CheckContext): Promise<PageUrlResult> {
 
   // 1. Try llms.txt links from cached results (if llms-txt-exists ran)
   const cachedUrls = await getUrlsFromCachedLlmsTxt(ctx);
-  const scopedCachedUrls = refineUrls(filterByPathPrefix(cachedUrls, filterBase));
-  if (scopedCachedUrls.length > 0) return { urls: scopedCachedUrls, warnings };
+  let llmsTxtUrls = refineUrls(filterByPathPrefix(cachedUrls, filterBase));
 
   // 2. Try fetching llms.txt directly (standalone mode, llms-txt-exists didn't run)
-  if (!ctx.previousResults.has('llms-txt-exists')) {
+  if (llmsTxtUrls.length === 0 && !ctx.previousResults.has('llms-txt-exists')) {
     const fetchedUrls = await fetchLlmsTxtUrls(ctx);
-    const scopedFetchedUrls = refineUrls(filterByPathPrefix(fetchedUrls, filterBase));
-    if (scopedFetchedUrls.length > 0) return { urls: scopedFetchedUrls, warnings };
+    llmsTxtUrls = refineUrls(filterByPathPrefix(fetchedUrls, filterBase));
+  }
+
+  if (llmsTxtUrls.length > 0) {
+    sources.push('llms-txt');
+
+    // If llms.txt meets the requested sample size, no need for sitemap
+    if (llmsTxtUrls.length >= ctx.options.maxLinksToTest) {
+      return { urls: llmsTxtUrls, warnings, sources };
+    }
+
+    // llms.txt is thin — try sitemap to fill the gap
+    const sitemapUrls = await getUrlsFromSitemap(ctx, warnings, { pathFilterBase: filterBase });
+    if (sitemapUrls.length > 0) {
+      sources.push('sitemap');
+      return { urls: mergeUrlSets(llmsTxtUrls, sitemapUrls), warnings, sources };
+    }
+
+    // Sitemap had nothing; return llms.txt URLs alone
+    return { urls: llmsTxtUrls, warnings, sources };
   }
 
   // 3. Try sitemap (path, locale, and version filtering applied inside)
   const sitemapUrls = await getUrlsFromSitemap(ctx, warnings, { pathFilterBase: filterBase });
-  if (sitemapUrls.length > 0) return { urls: sitemapUrls, warnings };
+  if (sitemapUrls.length > 0) {
+    sources.push('sitemap');
+    return { urls: sitemapUrls, warnings, sources };
+  }
 
   // 4. Fallback
-  return { urls: [ctx.baseUrl], warnings };
+  sources.push('fallback');
+  return { urls: [ctx.baseUrl], warnings, sources };
 }
 
 export interface SampledPages {
@@ -795,6 +841,8 @@ export interface SampledPages {
   warnings: string[];
   /** When curated pages have tags, maps page URL to tag label. */
   urlTags?: Record<string, string>;
+  /** Which discovery methods contributed to the page URL set. */
+  sources?: DiscoverySource[];
 }
 
 /**
@@ -888,6 +936,12 @@ export async function discoverAndSamplePages(ctx: CheckContext): Promise<Sampled
     }
   }
 
-  ctx._sampledPages = { urls, totalPages, sampled, warnings: discovery.warnings };
+  ctx._sampledPages = {
+    urls,
+    totalPages,
+    sampled,
+    warnings: discovery.warnings,
+    sources: discovery.sources,
+  };
   return ctx._sampledPages;
 }
diff --git a/src/runner.ts b/src/runner.ts
index d8a509b..af66165 100644
--- a/src/runner.ts
+++ b/src/runner.ts
@@ -125,6 +125,7 @@ export async function runChecks(
   };
 
   const urlTags = ctx._sampledPages?.urlTags;
+  const discoverySources = ctx._sampledPages?.sources;
 
   return {
     url: baseUrl,
@@ -133,5 +134,6 @@ export async function runChecks(
     results,
     summary,
     ...(urlTags && { urlTags }),
+    ...(discoverySources && { discoverySources }),
   };
 }
diff --git a/src/types.ts b/src/types.ts
index 9a6aebc..c7a5801 100644
--- a/src/types.ts
+++ b/src/types.ts
@@ -1,4 +1,4 @@
-import type { SampledPages } from './helpers/get-page-urls.js';
+import type { DiscoverySource, SampledPages } from './helpers/get-page-urls.js';
 
 export type CheckStatus = 'pass' | 'warn' | 'fail' | 'skip' | 'error';
 
@@ -154,6 +154,8 @@ export interface ReportResult {
   };
   /** When curated pages have tags, maps page URL to tag label. */
   urlTags?: Record<string, string>;
+  /** Which discovery methods contributed to the page URL set. */
+  discoverySources?: DiscoverySource[];
 }
 
 export interface AgentDocsConfig {
diff --git a/test/helpers/mock-sitemap-not-found.ts b/test/helpers/mock-sitemap-not-found.ts
new file mode 100644
index 0000000..2b30fda
--- /dev/null
+++ b/test/helpers/mock-sitemap-not-found.ts
@@ -0,0 +1,32 @@
+import { http, HttpResponse } from 'msw';
+import type { SetupServerApi } from 'msw/node';
+
+/**
+ * Register MSW handlers that return 404 for robots.txt and sitemap.xml
+ * at the given base URL's origin (and subpath if present).
+ *
+ * Call this after creating a test context with llms.txt content so that
+ * the discovery fallback (thin llms.txt → try sitemap) fails fast
+ * instead of timing out on unmocked test domains.
+ */
+export function mockSitemapNotFound(server: SetupServerApi, baseUrl: string): void {
+  const parsed = new URL(baseUrl);
+  const handlers = [
+    http.get(`${parsed.origin}/robots.txt`, () => new HttpResponse('', { status: 404 })),
+    http.get(`${parsed.origin}/sitemap.xml`, () => new HttpResponse('', { status: 404 })),
+  ];
+  const subpath = parsed.pathname.replace(/\/$/, '');
+  if (subpath && subpath !== '') {
+    handlers.push(
+      http.get(
+        `${parsed.origin}${subpath}/sitemap.xml`,
+        () => new HttpResponse('', { status: 404 }),
+      ),
+      http.get(
+        `${parsed.origin}${subpath}/sitemap-index.xml`,
+        () => new HttpResponse('', { status: 404 }),
+      ),
+    );
+  }
+  server.use(...handlers);
+}
diff --git a/test/integration/check-pipeline.test.ts b/test/integration/check-pipeline.test.ts
index 27b6869..12cbb01 100644
--- a/test/integration/check-pipeline.test.ts
+++ b/test/integration/check-pipeline.test.ts
@@ -11,6 +11,7 @@ import { http, HttpResponse } from 'msw';
 import { setupServer } from 'msw/node';
 import { runChecks } from '../../src/runner.js';
 import '../../src/checks/index.js';
+import { mockSitemapNotFound } from '../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -51,10 +52,16 @@ function setupSite(
     http.get(`http://${host}/docs/llms.txt`, () => new HttpResponse(null, { status: 404 })),
   );
 
+  // Sitemap discovery: return 404 so the fallback doesn't time out
+  handlers.push(
+    http.get(`http://${host}/robots.txt`, () => new HttpResponse('', { status: 404 })),
+    http.get(`http://${host}/sitemap.xml`, () => new HttpResponse('', { status: 404 })),
+  );
+
   const defaultCacheHeaders = opts.cacheControl ? { 'Cache-Control': opts.cacheControl } : {};
 
   for (const page of opts.pages) {
-    // HTML version
+    // HTML version (GET and HEAD)
     handlers.push(
       http.get(`http://${host}${page.path}`, ({ request }) => {
         const accept = request.headers.get('accept') ?? '';
@@ -72,9 +79,11 @@ function setupSite(
           { status: 200, headers: { 'Content-Type': 'text/html', ...defaultCacheHeaders } },
         );
       }),
+      // HEAD handler for llms-txt-links-resolve — mirrors GET status
+      http.head(`http://${host}${page.path}`, () => new HttpResponse(null, { status: 200 })),
     );
 
-    // .md URL
+    // .md URL (both /page.md and /page/index.md candidates from toMdUrls)
     if (page.md) {
       handlers.push(
         http.get(
@@ -85,10 +94,22 @@ function setupSite(
               headers: { 'Content-Type': 'text/markdown', ...defaultCacheHeaders },
             }),
         ),
+        http.get(
+          `http://${host}${page.path}/index.md`,
+          () =>
+            new HttpResponse(page.md!, {
+              status: 200,
+              headers: { 'Content-Type': 'text/markdown', ...defaultCacheHeaders },
+            }),
+        ),
       );
     } else {
       handlers.push(
         http.get(`http://${host}${page.path}.md`, () => new HttpResponse(null, { status: 404 })),
+        http.get(
+          `http://${host}${page.path}/index.md`,
+          () => new HttpResponse(null, { status: 404 }),
+        ),
       );
     }
 
@@ -566,11 +587,16 @@ describe('check pipeline: HTML fetch cache shared across checks', () => {
         'http://pipe-htmlcache.local/docs/guide.md',
         () => new HttpResponse(null, { status: 404 }),
       ),
+      http.get(
+        'http://pipe-htmlcache.local/docs/guide/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
       http.get(
         'http://pipe-htmlcache.local/docs/guide-afdocs-nonexistent-8f3a',
         () => new HttpResponse('Not Found', { status: 404 }),
       ),
     );
+    mockSitemapNotFound(server, 'http://pipe-htmlcache.local');
 
     const report = await runChecks('http://pipe-htmlcache.local', {
       checkIds: ['llms-txt-exists', 'page-size-html', 'tabbed-content-serialization'],
@@ -655,11 +681,16 @@ describe('check pipeline: auth-gate-detection → auth-alternative-access', () =
         'http://pipe-auth-llms.local/docs/page.md',
         () => new HttpResponse(null, { status: 404 }),
       ),
+      http.get(
+        'http://pipe-auth-llms.local/docs/page/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
       http.get(
         'http://pipe-auth-llms.local/docs/page-afdocs-nonexistent-8f3a',
         () => new HttpResponse('Not Found', { status: 404 }),
       ),
     );
+    mockSitemapNotFound(server, 'http://pipe-auth-llms.local');
 
     const report = await runChecks('http://pipe-auth-llms.local', {
       checkIds: ['llms-txt-exists', 'auth-gate-detection', 'auth-alternative-access'],
@@ -733,6 +764,14 @@ describe('check pipeline: auth-gate-detection → auth-alternative-access', () =
         'http://pipe-auth-md.local/docs/private.md',
         () => new HttpResponse(null, { status: 403 }),
       ),
+      http.get(
+        'http://pipe-auth-md.local/docs/public/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
+      http.get(
+        'http://pipe-auth-md.local/docs/private/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
       http.get(
         'http://pipe-auth-md.local/docs/public-afdocs-nonexistent-8f3a',
         () => new HttpResponse('Not Found', { status: 404 }),
@@ -742,6 +781,7 @@ describe('check pipeline: auth-gate-detection → auth-alternative-access', () =
         () => new HttpResponse('Not Found', { status: 404 }),
       ),
     );
+    mockSitemapNotFound(server, 'http://pipe-auth-md.local');
 
     const report = await runChecks('http://pipe-auth-md.local', {
       checkIds: [
@@ -802,12 +842,18 @@ describe('check pipeline: rendering-strategy → tabbed-content-serialization',
         'http://pipe-spa-tab.local/docs/guide.md',
         () => new HttpResponse(null, { status: 404 }),
       ),
+      http.get(
+        'http://pipe-spa-tab.local/docs/guide/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
       http.get(
         'http://pipe-spa-tab.local/docs/guide-afdocs-nonexistent-8f3a',
         () => new HttpResponse('Not Found', { status: 404 }),
       ),
     );
 
+    mockSitemapNotFound(server, 'http://pipe-spa-tab.local');
+
     const report = await runChecks('http://pipe-spa-tab.local', {
       checkIds: ['llms-txt-exists', 'rendering-strategy', 'tabbed-content-serialization'],
       requestDelay: 0,
@@ -866,12 +912,18 @@ describe('check pipeline: tabbed-content-serialization → section-header-qualit
         'http://pipe-tab-hdr.local/docs/install.md',
         () => new HttpResponse(null, { status: 404 }),
       ),
+      http.get(
+        'http://pipe-tab-hdr.local/docs/install/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
       http.get(
         'http://pipe-tab-hdr.local/docs/install-afdocs-nonexistent-8f3a',
         () => new HttpResponse('Not Found', { status: 404 }),
       ),
     );
 
+    mockSitemapNotFound(server, 'http://pipe-tab-hdr.local');
+
     const report = await runChecks('http://pipe-tab-hdr.local', {
       checkIds: ['llms-txt-exists', 'tabbed-content-serialization', 'section-header-quality'],
       requestDelay: 0,
@@ -921,6 +973,10 @@ describe('check pipeline: llms-txt-exists → llms-txt-links-markdown data flow'
             headers: { 'Content-Type': 'text/html' },
           }),
       ),
+      http.head(
+        'http://pipe-llms-md.local/docs/guide',
+        () => new HttpResponse(null, { status: 200 }),
+      ),
       // .md URL returns markdown
       http.get(
         'http://pipe-llms-md.local/docs/guide.md',
@@ -930,8 +986,18 @@ describe('check pipeline: llms-txt-exists → llms-txt-links-markdown data flow'
             headers: { 'Content-Type': 'text/markdown' },
           }),
       ),
+      http.head(
+        'http://pipe-llms-md.local/docs/guide.md',
+        () => new HttpResponse(null, { status: 200, headers: { 'Content-Type': 'text/markdown' } }),
+      ),
+      http.get(
+        'http://pipe-llms-md.local/docs/guide/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
     );
 
+    mockSitemapNotFound(server, 'http://pipe-llms-md.local');
+
     const report = await runChecks('http://pipe-llms-md.local', {
       checkIds: ['llms-txt-exists', 'llms-txt-links-markdown'],
       requestDelay: 0,
@@ -958,6 +1024,8 @@ describe('check pipeline: llms-txt-exists → llms-txt-links-markdown data flow'
       ),
     );
 
+    mockSitemapNotFound(server, 'http://pipe-llms-md-nollms.local');
+
     const report = await runChecks('http://pipe-llms-md-nollms.local', {
       checkIds: ['llms-txt-exists', 'llms-txt-links-markdown'],
       requestDelay: 0,
diff --git a/test/unit/checks/auth-gate-detection.test.ts b/test/unit/checks/auth-gate-detection.test.ts
index 8a81eda..92b5f08 100644
--- a/test/unit/checks/auth-gate-detection.test.ts
+++ b/test/unit/checks/auth-gate-detection.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -35,6 +36,7 @@ describe('auth-gate-detection', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
diff --git a/test/unit/checks/cache-header-hygiene.test.ts b/test/unit/checks/cache-header-hygiene.test.ts
index 5a0243c..122608b 100644
--- a/test/unit/checks/cache-header-hygiene.test.ts
+++ b/test/unit/checks/cache-header-hygiene.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -35,6 +36,7 @@ describe('cache-header-hygiene', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, `http://${host}`);
 
     return ctx;
   }
@@ -309,6 +311,7 @@ describe('cache-header-hygiene', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://chh-llms.local');
 
     const result = await check.run(ctx);
     expect(result.status).toBe('pass');
diff --git a/test/unit/checks/content-negotiation.test.ts b/test/unit/checks/content-negotiation.test.ts
index cfb201a..93d75fd 100644
--- a/test/unit/checks/content-negotiation.test.ts
+++ b/test/unit/checks/content-negotiation.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -30,6 +31,7 @@ describe('content-negotiation', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
@@ -190,6 +192,7 @@ describe('content-negotiation', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.details?.totalPages).toBe(5);
@@ -260,6 +263,7 @@ describe('content-negotiation', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.details?.sampled).toBe(true);
diff --git a/test/unit/checks/content-start-position.test.ts b/test/unit/checks/content-start-position.test.ts
index a96e3bb..c98f7fc 100644
--- a/test/unit/checks/content-start-position.test.ts
+++ b/test/unit/checks/content-start-position.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -30,6 +31,7 @@ describe('content-start-position', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
@@ -391,6 +393,7 @@ describe('content-start-position', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.details?.totalPages).toBe(5);
diff --git a/test/unit/checks/http-status-codes.test.ts b/test/unit/checks/http-status-codes.test.ts
index ebcee6f..4307328 100644
--- a/test/unit/checks/http-status-codes.test.ts
+++ b/test/unit/checks/http-status-codes.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -35,6 +36,7 @@ describe('http-status-codes', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
diff --git a/test/unit/checks/llms-txt-directive.test.ts b/test/unit/checks/llms-txt-directive.test.ts
index 3f443b9..e3aa584 100644
--- a/test/unit/checks/llms-txt-directive.test.ts
+++ b/test/unit/checks/llms-txt-directive.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -35,6 +36,7 @@ describe('llms-txt-directive', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
diff --git a/test/unit/checks/llms-txt-links-markdown.test.ts b/test/unit/checks/llms-txt-links-markdown.test.ts
index 19ed51d..f600e16 100644
--- a/test/unit/checks/llms-txt-links-markdown.test.ts
+++ b/test/unit/checks/llms-txt-links-markdown.test.ts
@@ -79,6 +79,7 @@ Just text, no links here.
           }),
       ),
       http.head('http://test.local/page1.md', () => new HttpResponse(null, { status: 404 })),
+      http.head('http://test.local/page1/index.md', () => new HttpResponse(null, { status: 404 })),
       http.head(
         'http://test.local/page2',
         () =>
@@ -88,6 +89,7 @@ Just text, no links here.
           }),
       ),
       http.head('http://test.local/page2.md', () => new HttpResponse(null, { status: 404 })),
+      http.head('http://test.local/page2/index.md', () => new HttpResponse(null, { status: 404 })),
     );
 
     const content = `# Test
diff --git a/test/unit/checks/markdown-code-fence-validity.test.ts b/test/unit/checks/markdown-code-fence-validity.test.ts
index 52b5446..d0cefaa 100644
--- a/test/unit/checks/markdown-code-fence-validity.test.ts
+++ b/test/unit/checks/markdown-code-fence-validity.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -53,6 +54,9 @@ describe('markdown-code-fence-validity', () => {
       message: llmsTxtFiles ? 'Found' : 'Not found',
       details: { discoveredFiles: llmsTxtFiles ?? [] },
     });
+    if (llmsTxtFiles) {
+      mockSitemapNotFound(server, 'http://test.local');
+    }
 
     return ctx;
   }
@@ -265,7 +269,12 @@ describe('markdown-code-fence-validity', () => {
             headers: { 'Content-Type': 'text/markdown' },
           }),
       ),
+      http.get(
+        'http://mcfv-standalone.local/docs/page/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
     );
+    mockSitemapNotFound(server, 'http://mcfv-standalone.local');
 
     // No dependency results set — standalone mode
     const ctx = createContext('http://mcfv-standalone.local', { requestDelay: 0 });
diff --git a/test/unit/checks/markdown-url-support.test.ts b/test/unit/checks/markdown-url-support.test.ts
index 15ed19a..2351aba 100644
--- a/test/unit/checks/markdown-url-support.test.ts
+++ b/test/unit/checks/markdown-url-support.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -35,6 +36,7 @@ describe('markdown-url-support', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
@@ -231,6 +233,7 @@ describe('markdown-url-support', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.details?.totalPages).toBe(5);
@@ -312,6 +315,7 @@ describe('markdown-url-support', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.details?.sampled).toBe(true);
diff --git a/test/unit/checks/page-size-html.test.ts b/test/unit/checks/page-size-html.test.ts
index 62c6d9e..58dea4d 100644
--- a/test/unit/checks/page-size-html.test.ts
+++ b/test/unit/checks/page-size-html.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -30,6 +31,7 @@ describe('page-size-html', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
@@ -95,6 +97,7 @@ describe('page-size-html', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.status).toBe('fail');
@@ -163,6 +166,7 @@ describe('page-size-html', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.details?.totalPages).toBe(5);
@@ -247,6 +251,7 @@ describe('page-size-html', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, 'http://test.local');
 
     const result = await check.run(ctx);
     expect(result.status).toBe('warn');
diff --git a/test/unit/checks/page-size-markdown.test.ts b/test/unit/checks/page-size-markdown.test.ts
index 165e205..211aa37 100644
--- a/test/unit/checks/page-size-markdown.test.ts
+++ b/test/unit/checks/page-size-markdown.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -30,6 +31,7 @@ describe('page-size-markdown', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     }
 
     return ctx;
@@ -163,7 +165,7 @@ describe('page-size-markdown', () => {
           }),
       ),
       http.get(
-        'http://ps-md-standalone.local.md',
+        'http://ps-md-standalone.local/.md',
         () => new HttpResponse('Not found', { status: 404 }),
       ),
       http.get(
@@ -196,7 +198,7 @@ describe('page-size-markdown', () => {
             headers: { 'Content-Type': 'text/html' },
           }),
       ),
-      http.get('http://ps-md-nomd.local.md', () => new HttpResponse('Not found', { status: 404 })),
+      http.get('http://ps-md-nomd.local/.md', () => new HttpResponse('Not found', { status: 404 })),
       http.get(
         'http://ps-md-nomd.local/index.md',
         () => new HttpResponse('Not found', { status: 404 }),
diff --git a/test/unit/checks/redirect-behavior.test.ts b/test/unit/checks/redirect-behavior.test.ts
index 8eebf50..fb02bc1 100644
--- a/test/unit/checks/redirect-behavior.test.ts
+++ b/test/unit/checks/redirect-behavior.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -35,6 +36,7 @@ describe('redirect-behavior', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
diff --git a/test/unit/checks/rendering-strategy.test.ts b/test/unit/checks/rendering-strategy.test.ts
index 75dda68..5239a47 100644
--- a/test/unit/checks/rendering-strategy.test.ts
+++ b/test/unit/checks/rendering-strategy.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -28,6 +29,7 @@ describe('rendering-strategy', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+    mockSitemapNotFound(server, `http://${domain}`);
     return ctx;
   }
 
diff --git a/test/unit/checks/tabbed-content-serialization.test.ts b/test/unit/checks/tabbed-content-serialization.test.ts
index 68d43a1..c34a3c0 100644
--- a/test/unit/checks/tabbed-content-serialization.test.ts
+++ b/test/unit/checks/tabbed-content-serialization.test.ts
@@ -5,6 +5,7 @@ import { createContext } from '../../../src/runner.js';
 import { getCheck } from '../../../src/checks/registry.js';
 import '../../../src/checks/index.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -30,6 +31,7 @@ describe('tabbed-content-serialization', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+      mockSitemapNotFound(server, 'http://test.local');
     } else {
       ctx.previousResults.set('llms-txt-exists', {
         id: 'llms-txt-exists',
diff --git a/test/unit/helpers/get-page-urls.test.ts b/test/unit/helpers/get-page-urls.test.ts
index 8bc6e12..6843453 100644
--- a/test/unit/helpers/get-page-urls.test.ts
+++ b/test/unit/helpers/get-page-urls.test.ts
@@ -17,6 +17,7 @@ import {
 import { MAX_SITEMAP_URLS } from '../../../src/constants.js';
 import { createContext } from '../../../src/runner.js';
 import type { DiscoveredFile } from '../../../src/types.js';
+import { mockSitemapNotFound } from '../../helpers/mock-sitemap-not-found.js';
 
 const server = setupServer();
 
@@ -582,6 +583,8 @@ describe('getPageUrls', () => {
         message: 'Found',
         details: { discoveredFiles: discovered },
       });
+
+      mockSitemapNotFound(server, baseUrl);
     } else {
       // Mark llms-txt-exists as having run (but failed) so getPageUrls
       // skips the direct llms.txt fetch and falls through to sitemap.
@@ -597,13 +600,14 @@ describe('getPageUrls', () => {
     return ctx;
   }
 
-  it('returns llms.txt links when available (no sitemap fetch)', async () => {
+  it('returns llms.txt links when available', async () => {
     const content = `# Docs\n> Summary\n## Links\n- [Page](http://test.local/docs/page): A page\n`;
     const ctx = makeCtx('http://test.local', content);
 
     const result = await getPageUrls(ctx);
     expect(result.urls).toEqual(['http://test.local/docs/page']);
     expect(result.warnings).toEqual([]);
+    expect(result.sources).toContain('llms-txt');
   });
 
   it('fetches and parses sitemap.xml when no llms.txt links', async () => {
@@ -632,6 +636,7 @@ describe('getPageUrls', () => {
       'http://sitemap-test.local/docs/intro',
       'http://sitemap-test.local/docs/guide',
     ]);
+    expect(result.sources).toEqual(['sitemap']);
   });
 
   it('handles sitemap index files (follows sub-sitemaps)', async () => {
@@ -708,6 +713,7 @@ describe('getPageUrls', () => {
     const ctx = makeCtx('http://empty-test.local');
     const result = await getPageUrls(ctx);
     expect(result.urls).toEqual(['http://empty-test.local']);
+    expect(result.sources).toEqual(['fallback']);
   });
 
   it('handles malformed sitemap XML gracefully', async () => {
@@ -1032,6 +1038,130 @@ describe('getPageUrls', () => {
     expect(result.warnings[0]).toContain('sitemap-docs.xml.gz');
   });
 
+  // ── Discovery source fallback: merge llms.txt + sitemap (#27) ──
+
+  it('falls back to sitemap when llms.txt has fewer URLs than maxLinksToTest', async () => {
+    const content = `# Docs\n## Links\n- [A](http://merge-test.local/docs/a): Page A\n- [B](http://merge-test.local/docs/b): Page B\n`;
+    const ctx = makeCtx('http://merge-test.local', content);
+    ctx.options.maxLinksToTest = 10;
+
+    // Register sitemap AFTER makeCtx so it takes precedence over the default 404 handlers
+    server.use(
+      http.get(
+        'http://merge-test.local/sitemap.xml',
+        () =>
+          new HttpResponse(
+            `<?xml version="1.0"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  <url><loc>http://merge-test.local/docs/c</loc></url>
+  <url><loc>http://merge-test.local/docs/d</loc></url>
+  <url><loc>http://merge-test.local/docs/e</loc></url>
+</urlset>`,
+            { status: 200, headers: { 'Content-Type': 'application/xml' } },
+          ),
+      ),
+    );
+
+    const result = await getPageUrls(ctx);
+    // llms.txt URLs come first, then sitemap fills the gap
+    expect(result.urls).toEqual([
+      'http://merge-test.local/docs/a',
+      'http://merge-test.local/docs/b',
+      'http://merge-test.local/docs/c',
+      'http://merge-test.local/docs/d',
+      'http://merge-test.local/docs/e',
+    ]);
+    expect(result.sources).toEqual(['llms-txt', 'sitemap']);
+  });
+
+  it('does not fall back to sitemap when llms.txt meets maxLinksToTest', async () => {
+    const content = `# Docs\n## Links\n- [A](http://no-merge.local/docs/a): A\n- [B](http://no-merge.local/docs/b): B\n- [C](http://no-merge.local/docs/c): C\n`;
+    const ctx = makeCtx('http://no-merge.local', content);
+    ctx.options.maxLinksToTest = 3;
+
+    const result = await getPageUrls(ctx);
+    expect(result.urls).toEqual([
+      'http://no-merge.local/docs/a',
+      'http://no-merge.local/docs/b',
+      'http://no-merge.local/docs/c',
+    ]);
+    expect(result.sources).toEqual(['llms-txt']);
+  });
+
+  it('deduplicates URLs when merging llms.txt and sitemap', async () => {
+    const content = `# Docs\n## Links\n- [A](http://dedup-merge.local/docs/a): A\n- [B](http://dedup-merge.local/docs/b): B\n`;
+    const ctx = makeCtx('http://dedup-merge.local', content);
+    ctx.options.maxLinksToTest = 10;
+
+    server.use(
+      http.get(
+        'http://dedup-merge.local/sitemap.xml',
+        () =>
+          new HttpResponse(
+            `<?xml version="1.0"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  <url><loc>http://dedup-merge.local/docs/a</loc></url>
+  <url><loc>http://dedup-merge.local/docs/b</loc></url>
+  <url><loc>http://dedup-merge.local/docs/c</loc></url>
+</urlset>`,
+            { status: 200, headers: { 'Content-Type': 'application/xml' } },
+          ),
+      ),
+    );
+
+    const result = await getPageUrls(ctx);
+    // Overlapping URLs should not be duplicated
+    expect(result.urls).toEqual([
+      'http://dedup-merge.local/docs/a',
+      'http://dedup-merge.local/docs/b',
+      'http://dedup-merge.local/docs/c',
+    ]);
+    expect(result.sources).toEqual(['llms-txt', 'sitemap']);
+  });
+
+  it('applies path-prefix filtering when merging llms.txt and sitemap', async () => {
+    const content = `# Docs\n## Links\n- [A](http://merge-scope.local/docs/a): A\n- [Blog](http://merge-scope.local/blog/post): Blog\n`;
+    const ctx = makeCtx('http://merge-scope.local/docs', content);
+    ctx.options.maxLinksToTest = 10;
+
+    // Register sitemap AFTER makeCtx so it takes precedence over the default 404 handlers
+    server.use(
+      http.get(
+        'http://merge-scope.local/sitemap.xml',
+        () =>
+          new HttpResponse(
+            `<?xml version="1.0"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  <url><loc>http://merge-scope.local/docs/b</loc></url>
+  <url><loc>http://merge-scope.local/docs/c</loc></url>
+  <url><loc>http://merge-scope.local/blog/other</loc></url>
+</urlset>`,
+            { status: 200, headers: { 'Content-Type': 'application/xml' } },
+          ),
+      ),
+    );
+
+    const result = await getPageUrls(ctx);
+    // llms.txt /blog/post filtered out by path prefix, sitemap /blog/other filtered too
+    expect(result.urls).toEqual([
+      'http://merge-scope.local/docs/a',
+      'http://merge-scope.local/docs/b',
+      'http://merge-scope.local/docs/c',
+    ]);
+    expect(result.sources).toEqual(['llms-txt', 'sitemap']);
+  });
+
+  it('reports only llms-txt source when sitemap is empty during merge attempt', async () => {
+    const content = `# Docs\n## Links\n- [A](http://thin-empty.local/docs/a): A\n`;
+    const ctx = makeCtx('http://thin-empty.local', content);
+    ctx.options.maxLinksToTest = 10;
+    // makeCtx already mocks sitemap as 404 — no sitemap URLs to merge
+
+    const result = await getPageUrls(ctx);
+    expect(result.urls).toEqual(['http://thin-empty.local/docs/a']);
+    expect(result.sources).toEqual(['llms-txt']);
+  });
+
   // ── Progressive disclosure: walking aggregate .txt files ──
 
   it('walks aggregate .txt files linked from llms.txt (Cloudflare pattern)', async () => {
@@ -1179,6 +1309,8 @@ describe('getPageUrls', () => {
         'http://direct-llms.local/docs/llms.txt',
         () => new HttpResponse('Not found', { status: 404 }),
       ),
+      http.get('http://direct-llms.local/robots.txt', () => new HttpResponse('', { status: 404 })),
+      http.get('http://direct-llms.local/sitemap.xml', () => new HttpResponse('', { status: 404 })),
     );
 
     // No llms-txt-exists in previousResults → standalone mode
@@ -1188,6 +1320,7 @@ describe('getPageUrls', () => {
       'http://direct-llms.local/docs/intro',
       'http://direct-llms.local/docs/guide',
     ]);
+    expect(result.sources).toContain('llms-txt');
   });
 
   it('skips llms.txt with non-text content-type in standalone mode', async () => {
@@ -1675,6 +1808,9 @@ describe('discoverAndSamplePages', () => {
       message: 'Found',
       details: { discoveredFiles: discovered },
     });
+
+    mockSitemapNotFound(server, baseUrl);
+
     return ctx;
   }
 
@@ -1687,6 +1823,7 @@ describe('discoverAndSamplePages', () => {
     expect(result.totalPages).toBe(2);
     expect(result.sampled).toBe(false);
     expect(result.warnings).toEqual([]);
+    expect(result.sources).toContain('llms-txt');
   });
 
   it('samples down to maxLinksToTest when over limit', async () => {
diff --git a/test/unit/runner.test.ts b/test/unit/runner.test.ts
index b687eeb..2b557cc 100644
--- a/test/unit/runner.test.ts
+++ b/test/unit/runner.test.ts
@@ -202,7 +202,7 @@ describe('runner', () => {
             headers: { 'Content-Type': 'text/html' },
           }),
       ),
-      http.get('http://standalone.local.md', () => new HttpResponse('Not found', { status: 404 })),
+      http.get('http://standalone.local/.md', () => new HttpResponse('Not found', { status: 404 })),
       http.get(
         'http://standalone.local/index.md',
         () => new HttpResponse('Not found', { status: 404 }),
@@ -296,6 +296,23 @@ describe('runner', () => {
   });
 
   it('includes timestamp and url in report', async () => {
+    server.use(
+      http.get('http://meta.local/llms.txt', () => new HttpResponse(null, { status: 404 })),
+      http.get('http://meta.local/docs/llms.txt', () => new HttpResponse(null, { status: 404 })),
+      http.get('http://meta.local/robots.txt', () => new HttpResponse('', { status: 404 })),
+      http.get('http://meta.local/sitemap.xml', () => new HttpResponse('', { status: 404 })),
+      http.get(
+        'http://meta.local',
+        () =>
+          new HttpResponse('<html><body><h1>Home</h1></body></html>', {
+            status: 200,
+            headers: { 'Content-Type': 'text/html' },
+          }),
+      ),
+      http.get('http://meta.local.md', () => new HttpResponse(null, { status: 404 })),
+      http.get('http://meta.local/index.md', () => new HttpResponse(null, { status: 404 })),
+    );
+
     const report = await runChecks('http://meta.local', {
       checkIds: ['tabbed-content-serialization'],
       requestDelay: 0,
@@ -304,4 +321,37 @@ describe('runner', () => {
     expect(report.url).toBe('http://meta.local');
     expect(report.timestamp).toMatch(/^\d{4}-\d{2}-\d{2}T/);
   });
+
+  it('includes discoverySources in report when page discovery runs', async () => {
+    server.use(
+      http.get('http://sources.local/llms.txt', () =>
+        HttpResponse.text('# Docs\n## Links\n- [A](http://sources.local/docs/a): A\n'),
+      ),
+      http.get('http://sources.local/docs/llms.txt', () => new HttpResponse(null, { status: 404 })),
+      http.get('http://sources.local/robots.txt', () => new HttpResponse('', { status: 404 })),
+      http.get('http://sources.local/sitemap.xml', () => new HttpResponse('', { status: 404 })),
+      http.get(
+        'http://sources.local/docs/a',
+        () =>
+          new HttpResponse('<html><body><h1>A</h1></body></html>', {
+            status: 200,
+            headers: { 'Content-Type': 'text/html' },
+          }),
+      ),
+      http.get('http://sources.local/docs/a.md', () => new HttpResponse(null, { status: 404 })),
+      http.get(
+        'http://sources.local/docs/a/index.md',
+        () => new HttpResponse(null, { status: 404 }),
+      ),
+    );
+
+    // markdown-url-support triggers discoverAndSamplePages, which populates sources
+    const report = await runChecks('http://sources.local', {
+      checkIds: ['llms-txt-exists', 'markdown-url-support'],
+      requestDelay: 0,
+    });
+
+    expect(report.discoverySources).toBeDefined();
+    expect(report.discoverySources).toContain('llms-txt');
+  });
 });