## Add Frame

`streams.add_frame(strnamespace, StreamAddFrameParams**kwargs)  -> StreamAddFrameResponse`

**post** `/v1/streams/{namespace}/frames`

Ingest a single livestream frame as an image data URI plus relative timestamp in milliseconds. The stream does not need to be created beforehand. Only image/jpeg, image/png, and image/webp are accepted. The decoded bytes must match the declared mime type, decode as a valid image, stay within 5 MB, and stay within 4096x4096 / 16,000,000 total pixels. Timestamps are relative milliseconds from stream start and must be unique and strictly increasing within a stream. Duplicate or out-of-order frame timestamps are accepted at the API boundary but ignored by downstream processing. This method returns immediately after the frame has been accepted for processing.

### Parameters

- `namespace: str`

  The namespace to ingest stream data into

- `image_data_uri: str`

  A data URI containing a base64-encoded frame image. Only image/jpeg, image/png, and image/webp are accepted. The decoded bytes must match the declared mime type, decode as a valid image, stay within 5 MB, and stay within 4096x4096 / 16,000,000 total pixels.

- `stream_id: str`

- `timestamp: int`

### Returns

- `class StreamAddFrameResponse: …`

  Stream ingestion request accepted for asynchronous processing.

  - `request_id: str`

  - `status: Literal["accepted"]`

    - `"accepted"`

### Example

```python
import os
from safetykit import Safetykit

client = Safetykit(
    api_key=os.environ.get("SAFETYKIT_API_KEY"),  # This is the default and can be omitted
)
response = client.streams.add_frame(
    namespace="namespace",
    image_data_uri="data:image/webp;base64,UklGRiIAAABXRUJQVlA4IBYAAAAwAQCdASoQABAAPm02mUmkIyIh...",
    stream_id="YOUR_STREAM_ID",
    timestamp=1000,
)
print(response.request_id)
```