I use the typical ChatGPT and Gemini LLMs a lot but to be honest, I am still waiting for a proper agent, something that can take over more complex tasks, maybe spawning sub-agents along the way.

Just today I tried to transcribe a short audio file and had to Google for a converter mp4->mp3 first, use it, then Google for a free transcriber that doesn't require registration and finally save the text which was difficult on mobile.

Does anyone know any proper AI Agents that are usable for more general use cases than just text replies?