Hey HN – I built this after realizing I was just a human copy-paste machine between users reporting bugs and Claude Code fixing them.
The flow: user speaks or types feedback → AI conversation gathers context → triage decides if actionable → coding agent clones your repo in a sandbox and implements the fix → PR appears in GitHub.
Free tier available, no credit card required to try it. Would love feedback on the voice UX especially.
Happy to answer questions about the architecture (Anthropic Claude Agent SDK, Vercel Sandbox, ElevenLabs for voice).
