Where are you getting these limitations from? They’re not in that article, and I went to the project’s page to double check and they’re not there either.

Connect any ACP-compatible agent or any model with an OpenAI-compatible API

At this point that’s basically anything. Including all the popular open frameworks fro running local AIs.

Automate workflows and recurring tasks: Completely removes your ability to make decisions and understand what is happening.

What? This is like setting a cron job. Does cron remove your ability to make decisions or understand what is happening?

Work seamlessly across devices with native applications for Windows, macOS, Linux, iOS, and Android: Until we decide it doesn’t, or maybe it will only be window.

It’s open source, like the other projects Mozilla maintains. Do you apply this “they could take it away from us at any time!” Concern to Firefox as well?

Maintain security with self-hosted deployment, optional end-to-end encryption, and device-level access controls: While allowing us to monitor your whole work flow remotely and monetize everything you know.

Any source for this? Seriously, I know there’s a lot of anti-AI sentiment around here but you’re hallucinating worse than Gemini.