diff --git a/memory/2026-03-04-voice-workflow-skill-invocation.md b/memory/2026-03-04-voice-workflow-skill-invocation.md
new file mode 100644
index 000000000..cbdcfd358
--- /dev/null
+++ b/memory/2026-03-04-voice-workflow-skill-invocation.md
@@ -0,0 +1,5 @@
+# 2026-03-04 Session Notes
+
+## Voice Workflow Skill Invocation
+
+Session began with a request to use the `/voice-workflow` skill. The skill provides transcription and text-to-speech functionality for voice message handling, using pywhispercpp for transcription (base.en model) and qwen3-tts-12hz-1.7b-base for TTS responses. However, no audio file was attached to the invocation, so a clarification was requested. Session ended without further action.
\ No newline at end of file