VOD frame extraction
Uploads video/audio to a BytePlus VOD space (from a local file or a public URL) and returns a vid://... reference. For media already in VOD, submits Snapshot tasks (StartExecution -> Operation.Task.Type: Snapshot) for frame extraction.
Product scope
| Aspect | Behaviour |
|---|---|
| Input | Vid or DirectUrl (JSON field video) |
| Extraction strategy | Default: specified time at 0 ms. Supported: specified time, fixed interval, specified frames, scene-change detection. |
| Target image size | Default resolution: 720p because the API Snapshot Target requires a resolution. Optional scale_long / scale_short. |
| Sprite image | Optional sprite / sprite_config. |
| Output index mode | Optional output_mode: Files or Index. |
| Advanced API fields | Use snapshot for complete passthrough or snapshot_options to deep-merge extra fields into generated Snapshot. |
If the user does not specify a strategy, use specified time at 0 ms. If they ask for multiple thumbnails but do not provide times, ask for the timestamps or use fixed interval only when they explicitly request evenly-spaced extraction.
Prerequisites
- Environment variables (required; optionally place a
.envin the working directory — scripts load it automatically):BYTEPLUS_ACCESSKEY— BytePlus Access KeyBYTEPLUS_SECRETKEY— BytePlus Secret KeyVOD_SPACE_NAME— VOD space name
- Environment template: see
scripts/env.md. - Execution: examples use
uv run python ...(python scripts/...works if deps are installed).
Workflow overview
Upload pipeline (local file):
[S1_APPLY] ApplyUploadInfo -> TOS upload address + SessionKey
[S2_TOS] PUT file to TOS (direct or chunked)
[S3_COMMIT] CommitUploadInfo -> Vid
Output: { Vid, Source, PlayURL, FileName, SpaceName, SourceUrl }
Upload pipeline (URL):
[S1_UPLOAD] Submit URL upload job (UploadMediaByUrl) -> JobId
[S2_POLL] Poll QueryUploadTaskInfo -> Vid
Output: { Vid, Source, PlayURL, FileName, SpaceName, SourceUrl, JobId }
Snapshot pipeline:
[S3_SNAPSHOT] Submit frame extraction task (StartExecution / Task.Type Snapshot) -> RunId
[S4_POLL] Poll GetExecution -> output snapshot files / raw Snapshot output
Output: { Status, SpaceName, ImageUrls[], Snapshot }
Quick self-check
Before running any script:
.envor env vars containBYTEPLUS_ACCESSKEY,BYTEPLUS_SECRETKEY, andVOD_SPACE_NAME.
Pick the pipeline from user intent:
| User intent | Pipeline | Entry script |
|---|---|---|
| Upload video to VOD | Upload | scripts/upload.py |
| Extract frames / thumbnails | Frame extraction | scripts/snapshot.py |
S1_UPLOAD & S2_POLL: upload and obtain Vid
Run from the Skill root directory (byted-byteplus-vod-frame-extraction/):
uv run python scripts/upload.py "/path/to/video.mp4" [space_name]
uv run python scripts/upload.py "https://example.com/video.mp4" [space_name]
- First argument: local file path or public
http:///https://URL. - Second argument (optional): space name; if omitted,
VOD_SPACE_NAMEis used. - Paths and URLs must include a file extension.
On success, preserve Source (vid://...) for downstream processing.
S3_SNAPSHOT & S4_POLL: frame extraction
Run from the Skill root directory (byted-byteplus-vod-frame-extraction/):
# Default: first frame at 0 ms, 720p
uv run python scripts/snapshot.py '{"type":"Vid","video":"v0310abc"}'
# Three exact timestamps in milliseconds
uv run python scripts/snapshot.py '{"type":"Vid","video":"vid://v0d225gxxx","strategy":"specified_time","times":[0,5000,10000],"resolution":"720p"}' production_space
# Every 3 seconds
uv run python scripts/snapshot.py '{"type":"Vid","video":"v0310abc","strategy":"interval","interval_ms":3000}'
# Scene-change snapshots
uv run python scripts/snapshot.py '{"type":"Vid","video":"v0310abc","strategy":"scene_change","threshold":0.1}'
uv run python scripts/snapshot.py @params.json
# Resume after timeout
uv run python scripts/poll_execution.py '<RunId>' [space_name]
Parameter reference
| Parameter | Type | Required | Description |
|---|---|---|---|
type |
string | no | Vid or DirectUrl. Default Vid. |
video |
string | yes | Vid or VOD FileName; vid:// / directurl:// prefixes are stripped automatically. |
strategy |
string/object | no | specified_time, interval, specified_frames, or scene_change. Default specified_time. If object, used directly as API Strategy. |
times |
integer/array | no | Millisecond offsets for specified_time. Default [0]. |
interval_ms |
integer | for interval | Millisecond interval for fixed interval snapshots. |
frames |
integer array | for specified frames | Frame indexes for specified_frames; 0 means first frame and -1 means last frame. |
threshold |
float | no | Scene-change threshold in [0, 1], default 0.1. |
resolution |
string | no | Default 720p. Allowed: 240p, 360p, 480p, 720p, 1080p. |
scale_long / scale_short |
integer | no | Long/short output image edge, [0, 4096]. |
sprite / sprite_config |
boolean/object | no | Sprite image config. Object is passed directly as SpriteConfig. |
output_mode |
string | no | Files or Index, maps to IndexOption.Mode. |
snapshot |
object | no | Complete API Snapshot object passthrough. |
snapshot_options |
object | no | Advanced fields deep-merged into generated Snapshot. |
Agent prompting
Ask for timestamps, interval, frame numbers, or scene-change detection only when the user's intent is ambiguous. Use conversational wording: “which frames or timestamps should I extract?” rather than raw API field names. If the user asks for a simple cover/thumbnail, use the default first-frame snapshot.
Output format
On success, one JSON object is printed to stdout:
{
"Status": "Success",
"SpaceName": "my_space",
"ImageUrls": [
{
"FileId": "...",
"Vid": "",
"DirectUrl": "path/to/snapshot.jpg",
"Source": "directurl://path/to/snapshot.jpg",
"Url": "https://example.cdn.com/...",
"Raw": {}
}
],
"VideoUrls": [],
"AudioUrls": [],
"Texts": [],
"Snapshot": {}
}
ImageUrls[].Url: playable / downloadable when signing succeeds for the space.Snapshot: raw API output fromOutput.Task.Snapshot.
Timeout handling
{
"error": "Polling timed out (360 attempts × 5s); the job is still processing",
"resume_hint": {
"description": "The job has not finished yet; resume polling with the command below",
"command": "uv run python scripts/poll_execution.py '<RunId>' [space_name]"
}
}
Environment variables
| Name | Description | Required |
|---|---|---|
BYTEPLUS_ACCESSKEY |
BytePlus Access Key | Yes |
BYTEPLUS_SECRETKEY |
BytePlus Secret Key | Yes |
VOD_SPACE_NAME |
VOD space name | Yes (or via CLI argument) |
VOD_POLL_INTERVAL |
Polling interval (seconds, default 5) | No |
VOD_POLL_MAX |
Maximum polling attempts (default 360) | No |
VOD_URL_EXPIRE_MINUTES |
Signed URL expiry (minutes, default 60) | No |
VOD_PLAY_DOMAIN |
Force a specific playback domain (optional, highest priority) | No |
VOD_HOST |
Override VOD OpenAPI hostname (optional) | No |
TOS_UPLOAD_CONNECT_TIMEOUT |
TOS upload connect timeout in seconds (default 5) | No |
TOS_UPLOAD_READ_TIMEOUT |
TOS upload read timeout in seconds (default 600) | No |
References
- StartExecution / OperationTaskSnapshot
- snapshot parameter reference
- API:
ApplyUploadInfo(2023-01-01) - API:
CommitUploadInfo(2023-01-01) - API:
UploadMediaByUrl(2023-01-01) - API:
QueryUploadTaskInfo(2023-01-01) - API:
StartExecution(2025-07-01) - API:
GetExecution(2025-07-01)