fix: address PR heygen-com#339 review — 7 issues

1. Catalog failure safety: warn when catalog is empty or throws, so
   capture doesn't silently produce zero images
2. Dead path: extract-audio-data.py → skills/gsap/scripts/ (was
   skills/hyperframes/scripts/)
3. --json fonts compat: emit both `fonts` (string[]) and
   `fontsDetailed` (FontToken[]) to avoid breaking external consumers
4. Restore .cursorrules writing alongside AGENTS.md + CLAUDE.md
5. .gitignore: remove over-broad `projects/` and `videos/` entries,
   keep scoped `cursor-tests/` and `launch-video*/`
6. agentPromptGenerator: mark unused params as reserved with comments,
   remove _animations from buildPrompt
7. Cookie filter: threshold 20 → 8 chars to preserve footer copy like
   "© 2026 Stripe" (16 chars) and "Privacy & Terms" (15 chars)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: ukimsanov

.gitignore

@@ -79,10 +79,8 @@ captures/
 *-tour/
 *-brand/
 
-# Local regression-test scratch directories (website-to-hyperframes runs).
-# These hold per-site captures, SCRIPT/STORYBOARD/DESIGN.md, compositions,
-# narration audio, rendered MP4s — all reproducible, never commit.
+# Local regression-test and video project scratch directories.
+# These hold per-site captures, compositions, narration audio, rendered MP4s.
 cursor-tests/
 basecamp-video/
-projects/
-videos/
+launch-video*/

packages/cli/src/capture/agentPromptGenerator.ts

@@ -20,30 +20,22 @@ export function generateAgentPrompt(
   outputDir: string,
   url: string,
   tokens: DesignTokens,
-  animations: AnimationCatalog | undefined,
+  _animations: AnimationCatalog | undefined, // reserved for future animation summary
   hasScreenshot: boolean,
   hasLottie?: boolean,
   hasShaders?: boolean,
-  _catalogedAssets?: CatalogedAsset[],
+  _catalogedAssets?: CatalogedAsset[], // reserved for future asset inventory
   detectedLibraries?: string[],
 ): void {
-  const prompt = buildPrompt(
-    url,
-    tokens,
-    animations,
-    hasScreenshot,
-    hasLottie,
-    hasShaders,
-    detectedLibraries,
-  );
+  const prompt = buildPrompt(url, tokens, hasScreenshot, hasLottie, hasShaders, detectedLibraries);
   writeFileSync(join(outputDir, "AGENTS.md"), prompt, "utf-8");
   writeFileSync(join(outputDir, "CLAUDE.md"), prompt, "utf-8");
+  writeFileSync(join(outputDir, ".cursorrules"), prompt, "utf-8");
 }
 
 function buildPrompt(
   url: string,
   tokens: DesignTokens,
-  _animations: AnimationCatalog | undefined,
   hasScreenshot: boolean,
   hasLottie?: boolean,
   hasShaders?: boolean,

packages/cli/src/capture/assetDownloader.ts

@@ -57,6 +57,8 @@ export async function downloadAssets(
   }
 
   // 3. Images — use the catalog as the single source of truth (highest resolution, deduplicated)
+  // If the catalog is empty, asset download produces zero images — this is surfaced as a warning
+  // so the capture doesn't silently produce a half-empty dataset.
   const imageUrls: { url: string; isPoster: boolean }[] = [];
 
   if (catalogedAssets && catalogedAssets.length > 0) {

packages/cli/src/capture/contentExtractor.ts

@@ -134,9 +134,10 @@ export async function extractVisibleText(page: Page): Promise<string> {
         if (style.display === 'none' || style.visibility === 'hidden' || style.opacity === '0') continue;
         var tag = el.tagName.toLowerCase();
         if (tag === 'script' || tag === 'style' || tag === 'noscript') continue;
-        // Skip short text inside nav/footer (catches nav links, cookie consent)
+        // Skip very short text inside nav/footer (catches single-word nav links)
+        // Threshold is 8 chars to preserve footer copy like "© 2026 Stripe" (16 chars)
         var inNavOrFooter = el.closest('nav, footer, [role="navigation"]');
-        if (inNavOrFooter && text.length < 20) continue;
+        if (inNavOrFooter && text.length < 8) continue;
         // Skip common cookie/consent patterns
         if (cookieRe.test(text)) continue;
         texts.push('[' + tag + '] ' + text);

packages/cli/src/capture/index.ts

@@ -339,8 +339,13 @@ export async function captureWebsite(
       const { catalogAssets } = await import("./assetCataloger.js");
       catalogedAssets = await catalogAssets(page1);
       progress("design", `${catalogedAssets.length} assets cataloged`);
+      if (catalogedAssets.length === 0) {
+        warnings.push(
+          "Asset catalog is empty — no images will be downloaded. The page may use non-standard image loading.",
+        );
+      }
     } catch (err) {
-      warnings.push(`Asset cataloging failed: ${err}`);
+      warnings.push(`Asset cataloging failed (no images will be downloaded): ${err}`);
     }
 
     // ── MUTATION phase: extractHtml modifies the live DOM (converts images to data URLs) ──

packages/cli/src/commands/capture.ts

@@ -115,7 +115,8 @@ export default defineCommand({
               screenshots: result.screenshots.length,
               assets: result.assets.length,
               detectedSections: result.tokens.sections.length,
-              fonts: result.tokens.fonts,
+              fonts: result.tokens.fonts.map((f) => f.family),
+              fontsDetailed: result.tokens.fonts,
               animations: result.animationCatalog?.summary,
               warnings: result.warnings,
             },

skills/website-to-hyperframes/references/techniques.md

@@ -367,7 +367,7 @@ for (var f = 0; f < AUDIO_DATA.totalFrames; f++) {
 Per-frame sampling is required — a single tween will not react. Use the extract script:
 
 ```bash
-python3 skills/hyperframes/scripts/extract-audio-data.py narration.wav --fps 30 --bands 16 -o audio-data.json
+python3 skills/gsap/scripts/extract-audio-data.py narration.wav --fps 30 --bands 16 -o audio-data.json
 ```
 
 Keep text/logo intensity subtle (≤5% scale, ≤30% glow) — audio-reactive motion on tiny elements reads as jitter. Bigger backgrounds can push to 10–30%.