ohad6k/VibeRaven

VibeRaven

AI got your app to demo. VibeRaven gets it to production.

Open the localhost launch console for AI-built apps before an agent says "ship it":

npx -y viberaven

Explicit UI and agent commands:

npx -y viberaven ui .
npx -y viberaven --agent-mode .
npx -y viberaven --verify .

VibeRaven runs the VibeRaven Production Protocol through a local-first open-source CLI/UI. The default command opens the localhost launch console; agent mode writes .viberaven/prp.json, .viberaven/gate-result.json, and supporting protocol artifacts so AI coding agents can keep operating until decision.status is not blocked and the gate is clear or a provider/user blocker remains.

Current npm live release: viberaven@1.1.10, @viberaven/cli@1.1.10, and @viberaven/mcp@1.1.8. The next CLI candidate is 1.1.11 for the refreshed Project Mission Control localhost UI.

If this repo helps, star it so other AI app builders can find the gate. Use Watch -> Custom -> Releases if you want release notifications.

VibeRaven public repo is the agent discovery and installation surface. Product source code and service internals live in a private repository.

Example scan output · What the gate checks · Full agent reference

Repositories: Public discovery → ohad6k/VibeRaven (this repo). Private product development → ohad6k/viberaven-dev (not public).

The localhost launch console is designed for vibe coders and coding agents: a startup screen with editable project icon, name, and version; a Project Mission Control status strip; "Can I launch?" as the top-level answer; provider actions for Supabase, Vercel, Stripe, GitHub, Sentry, Clerk, and PostHog; focused ag