Platforms
APPAI
Master control plane for the fleet.
One cockpit for 100 apps. APPAI orchestrates every app in ~/APPS — backlog → build → preview → push → deploy — so scaling the fleet doesn't mean scaling the chaos.
- Bulk build + deploy across the whole fleet from one dashboard
- Per-app dev ports, S3 buckets, and Vercel deployment status in one view
- Skill + agent registry maps every tool to the app it belongs to
- Apps registry (JSON) as the single source of truth for the fleet
Why
Managing 100 apps individually is a full-time job. APPAI makes fleet management a dashboard task — one build-all, one push-all, one status board, and a PBI backlog per app.
How
- PBI → build → preview → push → deploy, all orchestrated from the cockpit
- Per-app dev ports, S3 buckets, and Vercel deployments tracked in apps-registry.json
- Skill + agent registry ensures every Claude Code tool is wired to the right repo
Proof
- Apps in fleet today
- 13 → 100 target
- Build → deploy
- one command
- Source of truth
- apps-registry.json
APPAI — fleet cockpit loop
PBI · Build · Preview · Push · Deploy
Hover or tap a node to see details.
FAQ
- How many apps can APPAI handle?
- Designed for 100+. The registry is JSON-on-disk so it scales linearly with apps, not exponentially with cross-app links — adding the hundredth app costs the same as adding the tenth.
- Does it work on Windows / WSL2 / Mac?
- Yes to all three. Built on Node, Next.js, and standard Unix tooling — gh, git, vercel. Tested daily on WSL2 and macOS in real engagements.
- What happens when Vercel is down?
- APPAI cockpit runs locally, so the cockpit itself stays up. Deploy buttons hide automatically when Vercel can't be reached; everything else (status, push, build) keeps working.
In production
- 13 sites managed
Build, preview, push, deploy across all 13 production sites from one cockpit and one apps registry.
See it - JSON-on-disk source of truth
apps-registry.json drives every fleet verb — survives Vercel deploys, version-controlled, scriptable.
- Bulk verbs
build-all + push-all + deploy-all run concurrently with status piped to the run log and the /deploys UI.