dograh-hq/dograh — reverse-engineered prompt

Build me an open source voice AI platform that I can run myself, kind of like Vapi or Retell but without being locked into a hosted service.

I want a simple web dashboard where I can create an inbound or outbound voice bot, name it, describe what it should do, and then edit the conversation with a visual drag and drop workflow builder. I should be able to test the bot from the browser with a web call before connecting any phone provider.

Please include support for using built in test keys so it works right away, but also let me bring my own LLM, speech to text, and text to speech providers later. Add telephony support for services like Twilio, Vonage, Telnyx, or similar, including the ability to transfer a call to a human.

Make it Docker friendly so I can start it locally easily, with clear setup steps and basic docs. Look up current docs online if you need to.