diff --git a/internal/adapters/youtube/captions.go b/internal/adapters/youtube/captions.go
index 5797bdc..de87d88 100644
--- a/internal/adapters/youtube/captions.go
+++ b/internal/adapters/youtube/captions.go
@@ -21,6 +21,11 @@ const defaultPlayerBaseURL = "https://www.youtube.com"
 // returns caption baseUrls without a PoToken requirement (ADR-010).
 const androidUserAgent = "com.google.android.youtube/20.10.38 (Linux; U; Android 11) gzip"
 
+// browserUserAgent is required for the watch-page scrape: with an app UA YouTube
+// serves a page WITHOUT the embedded ytInitialPlayerResponse, so captionTracks
+// come back empty. A desktop-browser UA returns the player JSON with captions.
+const browserUserAgent = "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36"
+
 // maxCaptionBytes bounds a single timedtext/player body read.
 const maxCaptionBytes = 16 << 20 // 16 MiB
 
@@ -134,7 +139,7 @@ const ytInitialMarker = "ytInitialPlayerResponse"
 // embedded ytInitialPlayerResponse JSON. Any failure degrades to no tracks.
 func (a *Adapter) scrapeCaptionTracks(ctx context.Context, client *http.Client, videoID string) ([]captionTrack, error) {
 	body, status, err := a.httpGet(ctx, client, a.playerBaseURL+"/watch?v="+videoID, map[string]string{
-		"User-Agent": androidUserAgent,
+		"User-Agent": browserUserAgent,
 	})
 	if err != nil {
 		return nil, err