Record a walkthrough, import it, and your AI coworkers can reference it while they code. No context lost in translation.
How it works
You record. SageOx does the rest. Every recording goes through an extraction pipeline that turns video into structured, searchable artifacts your AI coworkers actually understand.
The pipeline runs automatically after upload. When it finishes, the extracted artifacts are committed to your Team Context repo — the shared knowledge base that every AI coworker on your team reads at the start of each session.
What gets extracted
Each recording produces a set of artifacts committed to your Team Context:
discussions/2026-03-20-ux-review/
├── transcript.vtt # timestamped speech with speaker labels
├── summary.json # chapters, decisions, action items
├── keyframes.json # frame images + vision descriptions
└── metadata.json # title, participants, durationYour AI coworkers consume these artifacts to understand what was discussed, what decisions were made, and what the UI looked like at the time. They don't watch the video — they read the structured output.
Smaller files = faster everything. 720p at 15fps is the sweet spot. Smaller files upload faster, transcribe faster, extract cleaner keyframes, and AI coworkers process the images more efficiently. Target ~1 MB per minute. No need for 4K — your AI coworker doesn't care about retina quality.
Getting started
| Guide | What you'll learn |
|---|---|
| Cap Setup (Recommended) | Optimal recording settings for AI extraction |
| Loom Import | Import existing Loom recordings |
| Upload via Web | Drag-and-drop upload in the browser |
| Import via CLI | Bulk import with ox import |
| Using in Coding Sessions | How AI coworkers consume your recordings |