🐳 Cloudsmith Datadog Agent Image Builder

This repository automates the build and publication of a Datadog Agent Docker image with the Cloudsmith integration pre-installed.

📦 Image published to Docker Hub → cloudsmith/cloudsmith-datadog-agent and Cloudsmith → cloudsmith/cloudsmith-datadog-agent

📌 Purpose

Datadog users who integrate Cloudsmith can now use a prebuilt Agent image with the datadog-cloudsmith integration bundled in. This reduces setup time and ensures consistent observability.

🚀 How It Works

This repo contains a GitHub Actions workflow that:

Watches the official Datadog integrations-extras repo for new Cloudsmith integration tags (cloudsmith-*)
Triggers a Docker build using the official Datadog Agent base image
Installs the datadog-cloudsmith integration at the corresponding version
Pushes the final image to Cloudsmith and Docker Hub

🛠 Usage

To run the image locally:

docker run -it --rm \
  -e DD_API_KEY=<your-api-key> \
  -e DD_SITE=datadoghq.com \
  -e DD_HOSTNAME=custom-host \
  -v ./conf.yaml:/etc/datadog-agent/conf.d/cloudsmith.d/conf.yaml \
  -v ./datadog.yaml:/etc/datadog-agent/datadog.yaml:ro \
  cloudsmith/cloudsmith-datadog-agent:<version>

Replace <version> with the appropriate tag (e.g., 1.0.0, 1.1.0, etc).
You can also mount the Cloudsmith config dynamically instead of copying it into the image.

📥 Tags Monitoring

The CI workflow runs daily (via cron) and monitors for new tags like cloudsmith-1.2.3.

When a new tag appears:

It triggers an automated Docker build
The new image is pushed to Docker Hub
.last-built-version is updated in this repo

📁 Project Structure

.
├── Dockerfile                   # Default build using Datadog’s official base image    
├── .last-built-version         # Tracks the most recently built tag  
├── README.md                   # GitHub repo README  
└── .github/  
    └── workflows/  
        └── build-cloudsmith-agent.yml   # CI workflow to auto-build and publish Docker image

👥 Maintainer

The image and workflow are maintained by Cloudsmith Customer Engineering team. Built for both public use and internal observability integrations.

If you’re using Cloudsmith to distribute your packages, this is the easiest way to monitor activity directly within your Datadog dashboards.