coder1317/ARIA ? reverse-engineered prompt
Reverse engineered prompt
Build me a local first autonomous AI assistant for software work that runs from the terminal and acts like a small team of agents, not just one chatbot.
I want it to research things with real sources, plan features, write code, review itself, debug, run tests, and keep improving from past runs. It should work fully on my laptop with Ollama, no API keys required by default, and it should run on Linux, macOS, and Windows. Please make it able to handle background tasks, remember goals and past work, keep an audit log of everything it does, and protect me from bad prompts or risky code.
I also want a simple API mode, Telegram support, browser based research for pages that need it, scheduled tasks, and a way to add outside tools later. Please include setup scripts so I can get it running easily, and make sure it can build and organize projects end to end in a way that feels reliable and safe. If you need to check current docs online, do that.
Are you gonna build this?
make sure you review the code using coderabbit