Example
Response:{
"count": 3,
"transcriptions": [
{
"title": "How to Build a Startup",
"duration": 1076.12,
"duration_formatted": "17:56",
"date": "2026-01-31 22:36",
"model_size": "tiny",
"md_path": "/Users/you/.augent/memory/transcriptions/How_to_Build_a_Startup.md",
"file_path": "/Users/you/Downloads/How to Build a Startup.webm"
}
],
"message": "Found 3 stored transcription(s)"
}
Notes
Titles are derived from the audio filename. Since
download_audio uses yt-dlp’s %(title)s template, downloaded videos are automatically named by their video title.Each stored transcription also has a markdown file at
~/.augent/memory/transcriptions/ that you can browse directly.