1. Pull RTSP stream from camera as described here.
2. Capture RTMP stream from RTMP encoder as described here.
3. Pull RTMP stream as client from another RTMP server as described here.
So your camera should act as RTSP source, RTMP encoder or RTMP server.
Also, if your camera act as a browser and can establish WebRTC connection, you can try to use raw Websocket API to publish stream, but it is probably a more complex way.
We have not tested Gstreamer yet. We plan to test it, but it may take a time.
If you can establish RTMP connection with WCS and start streaming, but WCS breaks the connection or does not play the stream, please hrepare a report as described here, including traffic dump, and send to firstname.lastname@example.org. We will check.
I think the tee command has to be used in gstreamer but i really can't figure out how to use it. Basicly should be possible tom get one stream and divide it into 2 streams directed to different queues. In theory doesn't seem so difficult but i scratched my head over it for weeks without any acceptable results.