VibeRaven

About

VibeRaven exists for the gap between 'the agent says it's done' and 'this is safe to deploy': AI-built apps routinely miss production essentials like auth wiring, service-role key exposure, webhook configuration, and monitoring. Running entirely locally with no login or telemetry, it connects the coding agent to the app's real context — stack and provider detection, release timeline, architecture map — and produces a gated ship-readiness verdict from offline checks covering auth/RLS, secret exposure, webhooks, and deploy readiness; a CI mode exits nonzero on blockers. Its agent cockpit drives Codex, Claude Code, or Gemini CLI from one UI with ask/approve/full access modes, and every artifact lives as plain markdown under .viberaven/, versioned by git and readable by any agent. Indie developers and small teams shipping AI-built apps use it as a pre-deploy gate; note the core product code is developed in a private repository, with the public repo carrying the CLI and agent integration surface.