Labrador

Labrador aims to make serverless infrastructure simple, repeatable, and manageable.

It uses JSON to define your cloud infrastructure — making it easy to version, review, and deploy serverless projects.

---

Terminology

Term	Definition
Project	The application you're building
Resource	An AWS resource (Lambda, S3, API Gateway, etc.)
Stage	A group of resources

---

Installation

Labrador is distributed as a standalone binary.

1. Download the latest release for your system.
2. Make it executable:

   chmod +x labrador

3. Move it to your PATH (for example):

   sudo mv labrador /usr/local/bin/labrador

After installation, you can verify it works by running:

labrador --version

---

Quickstart

Create an env file with AWS credentials:

# .env
AWS_ACCESS_KEY_ID=
AWS_SECRET_ACCESS_KEY=
AWS_REGION=

Initialize a new Labrador project:

labrador init --name my_project --env dev --output my_project.json

Add a stage to your project:

labrador add stage --project my_project.json --type s3 --name assets --output buckets.json

Inspect your infrastructure:

labrador inspect --project my_project.json --env-file .env --full

Deploy your infrastructure:

labrador deploy --project my_project.json --env-file .env

More than just S3.

Labrador can also scaffold and deploy Lambda functions and API Gateways:

labrador add stage --type lambda --help
labrador add stage --type api --help

Each stage type supports customizable configuration, sensible defaults, and full environment interpolation.

Examples

See /examples for a sample project deploying several lambda functions, and API gateway with several integrations and routes, and an S3 bucket. You'll need to provide one or more Role ARNs for the lambdas if you choose to deploy it.

---

AWS Credentials

You have two options for passing AWS credentials to Labrador:

1. Use global flags (--aws-access-key-id, --aws-secret-access-key, --aws-region)
2. Use an environment file defining AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_REGION

Note: Labrador will never implicitly read .env. You must use --env-file=.env if you want to use variables defined there. Labrador will implicitly read .labrador.env though.

---

Defining Infrastructure

Labrador organizes your infrastructure into projects and stages, with each configuration defined in a JSON file.

File	Defines
Project config	Defines your project and settings
Stage config	Defines one or more cloud resources

---

Project Configuration

A project configuration defines the overall project metadata, such as name, environment, and variables.

Example:

{
  "name": "my-project",
  "environment": "dev",
  "variables": {
    "config_files": "./stages",
    "version": "1.0"
  },
  "stages": []
}

---

Supported Services

Service	Create	Update	Delete
Lambda	✅	✅	✅
S3	✅	✅	✅
API Gateway	✅	✅	✅

---

Status

Labrador is currently in early development (v0.2.x).
While functional, Labrador is still early in development. Minor rough edges may exist as we continue to refine and expand it.

Feedback, issues, and contributions are welcome!