Features
The fastest way to get a single recording into SageOx. Drag, drop, done.
Upload a recording
- Navigate to your team's Discussions page
- Click Upload in the top right
- Drag your recording into the upload area (or click Browse Files)
- Add a descriptive title — this is how AI coworkers find your recording
That's it. The pipeline kicks off automatically.
Supported formats
| Category | Formats |
|---|---|
| Video | MP4, WebM |
| Audio | M4A, MP3, WAV, OGG/Opus, FLAC, AAC, WMA |
| Transcript | VTT, SRT, TXT, DOCX |
Limits. 500 MB per file, up to 20 files per batch. For the fastest processing, keep video files under 100 MB. See Cap Setup for optimal recording settings.
The web UI supports file upload only. To import directly from a Loom or Cap share URL, use ox import.
What happens next
After upload, SageOx automatically:
- Extracts audio and transcribes with speaker identification
- Detects scene changes and extracts representative keyframes
- Analyzes keyframe images for UI elements, code, and visual context
- Generates a structured summary with chapters, decisions, and action items
- Commits all artifacts to your Team Context repo
Your AI coworkers pick up these artifacts the next time they run ox agent prime. They don't watch the video — they read the structured transcript, reference the keyframe descriptions, and use the summary to understand decisions and context.
Learn more about using recordings in coding sessions.