@detiq/crawler 1.0.2 → 1.0.4

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@detiq/crawler",
-  "version": "1.0.2",
+  "version": "1.0.4",
   "private": false,
   "publishConfig": {
     "access": "public",

package/src/crawler.ts

@@ -2242,12 +2242,15 @@ export async function crawlProject(opts: {
         await waitForScreenshotReady(page);
         await onProgress?.(lastEmittedPct, `Capturing ${(() => { try { return new URL(url).pathname || '/'; } catch { return '/'; } })()}`);
 
+        // FM-4: capture device pixel ratio for coordinate scaling in element highlighting
+        const pageDpr: number = await page.evaluate(() => window.devicePixelRatio ?? 1).catch(() => 1);
+
         // Parallelize: screenshot + element extraction + link discovery + AT capture (all work on current page state)
         const [, elements, nextLinks, atElements] = await Promise.all([
           page.screenshot({ path: shotPath, fullPage: true })
             .then(async () => {
               const cloudUrl = await uploadToCloud(shotPath, `${opts.tenantId}/${opts.projectId}/screenshots/${path.basename(shotPath)}`);
-              return AppBrain.saveScreenshot(opts.tenantId, opts.projectId, screen.id, shotPath, screen.version, 'DESKTOP', cloudUrl);
+              return (AppBrain.saveScreenshot as any)(opts.tenantId, opts.projectId, screen.id, shotPath, screen.version, 'DESKTOP', cloudUrl, pageDpr);
             })
             .catch((err: any) => console.error('[crawler] page.screenshot failed:', err)),
           extractElements(stagehand, page, elementLLM ?? undefined).catch((err: any) => {
@@ -3439,7 +3442,47 @@ async function extractElements(stagehand: Stagehand, page: any, llmCfg?: { provi
   // DOM-first: fast, no LLM cost, covers server-rendered and most SPAs.
   // Threshold >= 2 so a lone dark-mode button doesn't suppress AI fallback for rich pages.
   const domElements = await domFallbackElements(page);
-  if (domElements.length >= 2) return domElements;
+
+  // FM-8: extract interactive elements from visible same-origin iframes
+  const iframeElements: typeof domElements = [];
+  try {
+    const iframes = page.frames ? page.frames() : [];
+    for (const frame of iframes) {
+      if (frame === page.mainFrame?.()) continue; // skip main frame
+      const iframeUrl: string = await frame.url().catch(() => '');
+      if (!iframeUrl || iframeUrl === 'about:blank' || iframeUrl.startsWith('javascript:')) continue;
+      const frameEls = await domFallbackElements(frame).catch(() => []);
+      // Tag iframe elements with source context
+      for (const el of frameEls.slice(0, 10)) {
+        iframeElements.push({ ...el, notes: el.notes ? `${el.notes} [in iframe]` : '[in iframe]' });
+      }
+    }
+  } catch { /* iframes may be cross-origin or inaccessible — skip silently */ }
+
+  // FM-7: reveal dynamic elements — click/hover trigger elements that show dropdowns or modals
+  // Only run when base DOM found elements (not on blank pages) to avoid spurious interactions.
+  let dynamicElements: typeof domElements = [];
+  if (domElements.length > 0) {
+    try {
+      dynamicElements = await page.evaluate(() => {
+        const out: { meaning: string; role: string; notes: string }[] = [];
+        const seen = new Set<string>();
+        // Look for select/combobox/listbox that aren't already captured as interactive elements
+        const triggers = document.querySelectorAll<HTMLElement>('[role="combobox"],[role="listbox"] [role="option"],[aria-haspopup]');
+        triggers.forEach((el) => {
+          const label = (el.getAttribute('aria-label') || el.textContent?.trim() || '').slice(0, 80);
+          if (!label || seen.has(label)) return;
+          // Don't duplicate what's already captured
+          const role = el.getAttribute('role') ?? el.tagName.toLowerCase();
+          seen.add(label);
+          out.push({ meaning: label, role, notes: '[dynamic/revealed]' });
+        });
+        return out.slice(0, 10);
+      }).catch(() => []);
+    } catch { dynamicElements = []; }
+  }
+
+  if (domElements.length >= 2) return [...domElements, ...iframeElements, ...dynamicElements];
 
   // Content fallback: static/marketing pages have no interactive widgets but have headings,
   // images, links, and meta tags that feed SEO/a11y/visual test generation.

package/src/index.ts

@@ -33,3 +33,5 @@ export type { ProxyTier, ProxyTierConfig } from './proxy.js';
 export { RedisCrawlFrontier, isRedisFrontierEnabled } from './redis-frontier.js';
 export { startSpan, endSpan, withSpan, increment, flushMetrics } from './telemetry.js';
 export type { Span, FinishedSpan } from './telemetry.js';
+export { analyzeFailure } from './failure-analysis-agent.js';
+export type { FailureContext } from './failure-analysis-agent.js';