diff --git a/commands/conversions/rich-text-clipboard-to-markdown.sh b/commands/conversions/rich-text-clipboard-to-markdown.sh
index 5bf7d5d1d..4e66d6379 100755
--- a/commands/conversions/rich-text-clipboard-to-markdown.sh
+++ b/commands/conversions/rich-text-clipboard-to-markdown.sh
@@ -6,7 +6,7 @@
 # @raycast.title Rich Text to Markdown
 # @raycast.author Adam Zethraeus
 # @raycast.authorURL https://github.com/adam-zethraeus
-# @raycast.description Convert rich text clipboard data to GitHub Flavored Markdown using Pandoc
+# @raycast.description Convert rich text clipboard data (preserving hyperlinked text as [label](url)) to GitHub Flavored Markdown using Pandoc. Tries the HTML pasteboard flavor first, falls back to RTF.
 #
 # @raycast.icon 📝
 #
@@ -20,4 +20,19 @@ if ! command -v pandoc &> /dev/null; then
 fi
 
 export LC_CTYPE=UTF-8
+
+# Prefer HTML: most modern web sources (browsers, Gmail, Google Docs, Notion,
+# Linear, etc.) place higher-fidelity HTML on the pasteboard than RTF, and it
+# preserves hyperlinked text as real anchor tags that pandoc renders as
+# [label](url) in markdown.
+html=$(osascript -e 'try' -e 'the clipboard as «class HTML»' -e 'on error' -e 'return ""' -e 'end try' 2>/dev/null \
+  | perl -ne 'chomp; next unless s/^«data HTML//; s/»$//; print pack("H*", $_)')
+
+if [ -n "$html" ]; then
+    printf '%s' "$html" | pandoc --from=html --to=gfm | pbcopy
+    exit 0
+fi
+
+# Fall back to RTF for sources that only expose rich text (some native apps,
+# older editors). This is the original behavior of the script.
 osascript -e 'the clipboard as «class RTF »' | perl -ne 'print chr foreach unpack("C*",pack("H*",substr($_,11,-3)))' | pandoc --from=rtf --to=gfm | pbcopy