Create a Generic Artifact

Convert custom scripts and tools into Bluebricks packages using Docker containers.

Prerequisites

• Bluebricks CLI installed

• Authenticated with Bluebricks (bricks login)

• Docker image published to an approved registry

• Basic knowledge of Docker and containers

Quick Start

1. Create package directory:

   mkdir my-generic-package
   cd my-generic-package

2. Create bricks.json manually:

   {
     "name": "my-generic-package",
     "version": "0.1.0",
     "description": "Custom executor package",
     "native": {
       "type": "generic",
       "path": "./src",
       "image": "python:3.11-slim",
       "command": ["python"],
       "args": ["/workspace/scripts/my_script.py"]
     },
     "props": {
       "input_param": {
         "type": "string",
         "description": "Input parameter"
       }
     },
     "outs": {
       "job_id": {
         "type": "string",
         "description": "Job ID of the generic package job"
       },
       "result": {
         "type": "string",
         "description": "Execution result"
       }
     }
   }

3. Create your script:

   # src/scripts/my_script.py
   import json
   
   # Read inputs
   with open('/workspace/vars.json', 'r') as f:
       vars = json.load(f)
   
   # Your custom logic here
   result = f"Processed: {vars['input_param']}"
   
   # Write outputs
   with open('/workspace/outputs.json', 'w') as f:
       json.dump({'result': result}, f)

4. Test and deploy:

   bricks run . --dry --props-file properties.json
   bricks run . --apply --props-file properties.json

When to Use Generic Artifacts

Use Generic artifacts when you need to:

• Run custom deployment scripts (Bash, Python, Node.js)

• Execute API calls or database migrations

• Integrate with tools not natively supported

• Build complex workflows with custom logic

• Orchestrate multi-step processes

How It Works

1. Package Structure: You create bricks.json manually with custom configuration

2. Execution: Bluebricks runs your container as a Kubernetes Job

3. Input: Props are passed as vars.json file mounted in container

4. Output: Container writes results to /workspace/outputs.json file. These outputs are registered as deployment outputs and can be referenced by downstream packages using Data.<package_id>.<output_key> syntax

5. State: Bluebricks tracks outputs and input hash in state.json

See also

• Generic Artifact Types -- common patterns and approved registries

• Container Configuration -- image, command, and environment setup

• Input/Output Handling -- vars.json and outputs.json contract

• Packages overview -- artifact types and packaging concepts