Server
On this page 5
stacks-server · Toolchain · model-invoked
The server itself, in development and production: configuration, middleware and startup.
When to reach for it
- Server configuration
- Server middleware
- Server startup
Covers
@stacksjs/server, storage/framework/server/.
Inside the skill
The sections an agent reads once the skill loads.
- Key Paths
- Source Files
- Server Entry Point (server/src/index.ts)
- Server Config (core/server/src/config.ts)
- Production Config (core/server/src/config-production.ts)
- Production Start (core/server/src/start.ts)
- Auto-Imports System (core/server/src/imports.ts)
- Base Controller (core/server/src/controllers/base.ts)
- Maintenance Mode (core/server/src/maintenance.ts)
- Docker Build Pipeline (server/build.ts)
- Dockerfile
- dev Script (server/dev)
- Environment Variables
- CLI Commands
- Gotchas
Where the code lives
- Core package:
storage/framework/core/server/src/(published as@stacksjs/server) - Server runtime:
storage/framework/server/(the actual Bun HTTP server + Docker build) - Server types:
storage/framework/core/types/src/server.ts - Ports types:
storage/framework/core/types/src/ports.ts - Package (core):
storage/framework/core/server/package.json - Package (runtime):
storage/framework/server/package.json
Using it
This one is model-invoked. Your agent reaches for it on its own when the task matches, and you can also call it by name:
/stacks-server
Source: stacks-server/SKILL.md.
Shadow it for one project with app/Skills/stacks-server/SKILL.md, then re-run
buddy setup:ai. See Writing your own.