# Services A Service in Hydron represents a single deployable component of your application. Most projects have one or more services — for example, a web application might have an API service, a frontend service, and a database service. ## How services are created Services are created in two ways: 1. **Automatically detected** — When you connect a Git repository, the AI analyzes your code and detects services based on the project structure, entry points, and configuration files 2. **Manually added** — You can add services through the chat or project settings ## Service properties Each service has the following properties: | Property | Description | |----------|-------------| | **Name** | A unique identifier within the project (e.g., `api-server`, `frontend`) | | **Type** | The kind of service (web server, worker, database, cache, etc.) | | **Source** | Where the code comes from (Git repository, Docker image) | | **Dockerfile** | The Docker configuration for building and running the service | | **Environment variables** | Configuration values injected at runtime | | **Ports** | Network ports the service exposes | | **Health check** | Endpoint and settings for monitoring service health | | **Dependencies** | Other services this one depends on | ## Viewing services Services are displayed in the **right sidebar** of the chat interface under the **Services** tab.  ## Accepting and rejecting services After code analysis, detected services are in a **pending** state. You need to review and either **accept** or **reject** each one: - **Accept** — Include this service in the infrastructure plan and deployment - **Reject** — Exclude this service from deployment Only accepted services are included when generating infrastructure plans. ## Environment variables Each service can have its own set of environment variables. These are injected into the Docker container at runtime. ### Adding environment variables You can add environment variables in several ways: - **Automatically detected** — The AI extracts variables referenced in your code during analysis - **Through the chat** — Tell the AI: "Set DATABASE_URL to postgres://..." - **Through the sidebar** — Click on a service and add variables in the configuration panel ### Variable enrichment Hydron can automatically enrich environment variables by detecting common patterns: - Database connection strings - API keys and secrets - Port configurations - Service discovery URLs (connecting your services to each other) ### Shared secrets When multiple services need the same environment variable (e.g., a shared database URL), Hydron can automatically configure cross-service references so you only define the value once. ## Dockerfiles Every service needs a Dockerfile to define how it's built and run. Hydron handles this in three ways: 1. **Existing Dockerfile** — If your repository already has a Dockerfile, Hydron uses it 2. **AI-generated** — The AI creates an optimized Dockerfile based on the detected framework 3. **Custom** — You can provide or edit a Dockerfile through the chat ### Editing Dockerfiles Ask the AI to modify the Dockerfile for any service: - "Use a multi-stage build for the frontend service" - "Change the Node.js version to 20-alpine" - "Add a health check command to the API Dockerfile" ## Service dependencies Services can depend on other services. For example: ~~~ api-server ──depends on──> postgres api-server ──depends on──> redis frontend ──depends on──> api-server ~~~ Dependencies affect: - **Startup order** — Dependent services start after their dependencies - **Networking** — Dependent services can reach their dependencies via internal DNS - **Health checks** — A service isn't considered healthy until its dependencies are healthy ## Locking services You can **lock** a service to prevent accidental changes. A locked service: - Cannot have its Dockerfile modified - Cannot have environment variables changed - Cannot be removed from the project This is useful once you've finalized a service's configuration and don't want it accidentally modified during further iterations.