Serverless CMS — Architecture & Technical Documentation
This section documents what the Serverless CMS is, the problems it solves, and how it is implemented on AWS. It is written for engineers, architects and technical evaluators who need to understand the system's design and technical specifications.
Looking for task-based guides instead? See the
user guide (day-to-day authoring) and the
developer guide (building components and sites).
For getting it running, see the installation guide.
Contents
# | Document | What it covers |
|---|---|---|
01 | What the CMS is, who it is for, and what it achieves | |
02 | The planes of the system, how requests flow, and the deployment topology | |
03 | Every deployed component (APIs, functions, stores, services) and its role | |
04 | The DynamoDB single-table design, key patterns and indexes | |
05 | The authenticated admin API and the public delivery API | |
06 | How pages are composed from components and published to static HTML | |
07 | Identity, authorization model, and per-request enforcement | |
08 | Runtimes, limits, tech stack, and the build/deploy workflow |
At a glance
- What — A multi-tenant, headless-capable CMS that lets teams create and operate many websites, author pages from reusable components, and publish them as fast static sites.
- How — 100% serverless on AWS: DynamoDB, S3, Lambda (Node.js 22), API Gateway, Cognito, CloudFront, SES/SNS. No servers to manage and pay-per-request scaling.
- Deployed by — The Serverless Maven Plugin, which turns a declarative JSON configuration into the full AWS resource graph.
The system is defined declaratively in src/main/serverless/config.json and the component files under src/main/serverless/components/. Those files are the source of truth for the architecture described here.