Skip to main content
POST
/
video-summaries
Create Video Summary
curl --request POST \
  --url https://api.bota.dev/v1/video-summaries \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "recording_id": "<string>",
  "media_ids": [
    "<string>"
  ],
  "max_keyframes": 123,
  "include_transcript_context": true
}
'
{
  "id": "vs_abc123",
  "recording_id": "rec_abc123",
  "status": "pending",
  "media_count": 3,
  "created_at": "2025-01-15T11:00:00Z"
}

Documentation Index

Fetch the complete documentation index at: https://docs.bota.dev/llms.txt

Use this file to discover all available pages before exploring further.

This feature is Coming Soon and not yet available.

​
Overview

Generates a visual summary from video media captured during a recording. The system identifies key frames, generates captions for each, and produces a timeline of visual highlights alongside the audio narrative. Useful for quickly reviewing long recordings — see the most important visual moments without watching the entire video.

​
Authentication

Requires a valid API key with video_summaries:write scope.

​
Request Body

​
recording_id
string
required
Recording ID with uploaded video media (rec_*)
​
media_ids
string[]
Specific video media IDs to summarize. If omitted, all video media is processed.
​
max_keyframes
integer
Maximum number of key frames to extract per video. Default: 10.
​
include_transcript_context
boolean
Whether to correlate key frames with transcript segments. Default: true.

​
Response

{
  "id": "vs_abc123",
  "recording_id": "rec_abc123",
  "status": "pending",
  "media_count": 3,
  "created_at": "2025-01-15T11:00:00Z"
}