Deploy: Docker

Deploy agents to Docker containers using praisonai deploy

Build and deploy agents as Docker containers using praisonai deploy run --type docker.

Quick Start - CLI

1. Install PraisonAI

   pip install praisonai
   
2. Set API Key

   export OPENAI_API_KEY="your-key"
   
3. Initialize Docker Config

   praisonai deploy init --file agents.yaml --type docker
   
4. Deploy to Docker

   praisonai deploy run --file agents.yaml --type docker
   

   Expected Output:

   🚀 Starting deployment...
   
   📦 Building Docker image: praisonai-app:latest
     • Base image: python:3.11-slim
     • Exposing ports: [8005]
   
   ✅ Deployment successful!
   🔗 URL: http://localhost:8005
   
   Metadata:
     • type: docker
     • image: praisonai-app:latest
     • container_id: abc123def456
   
5. Verify

   docker ps | grep praisonai
   curl http://localhost:8005/health
   

Quick Start - SDK

1. Create Deploy Script

   from praisonai.deploy import Deploy, DeployConfig, DeployType
   from praisonai.deploy.models import DockerConfig
   
   config = DeployConfig(
       type=DeployType.DOCKER,
       docker=DockerConfig(
           image_name="praisonai-app",
           tag="latest",
           base_image="python:3.11-slim",
           expose=[8005],
           push=False
       )
   )
   
   deploy = Deploy(config, "agents.yaml")
   result = deploy.deploy()
   
   print(f"Image: {result.metadata.get('image')}")
   print(f"Container: {result.metadata.get('container_id')}")
   
2. Run

   python deploy_docker.py
   

agents.yaml with Docker Config

framework: praisonai
topic: helpful assistant
roles:
  assistant:
    role: Assistant
    goal: Help users with their questions
    backstory: You are a helpful assistant
    tasks:
      help_task:
        description: Answer user questions
        expected_output: Helpful response

deploy:
  type: docker
  docker:
    image_name: "praisonai-app"
    tag: "latest"
    base_image: "python:3.11-slim"
    expose: [8005]
    push: false
    registry: null

Docker Config Options

Field	Type	Default	Description
image_name	string	praisonai-app	Docker image name
tag	string	latest	Docker image tag
base_image	string	python:3.11-slim	Base Docker image
expose	list[int]	[8005]	Ports to expose
registry	string	null	Docker registry URL
push	bool	false	Push image to registry
build_args	dict	null	Docker build arguments

Push to Registry

deploy:
  type: docker
  docker:
    image_name: "praisonai-app"
    tag: "v1.0.0"
    registry: "your-registry.com/your-org"
    push: true

praisonai deploy run --file agents.yaml --type docker

Expected Output:

🚀 Starting deployment...

📦 Building Docker image: your-registry.com/your-org/praisonai-app:v1.0.0
📤 Pushing to registry: your-registry.com/your-org

✅ Deployment successful!

Metadata:
  • image: your-registry.com/your-org/praisonai-app:v1.0.0
  • pushed: true

Check Status

praisonai deploy status --file agents.yaml

Stop Container

praisonai deploy destroy --file agents.yaml

Expected Output:

⚠️  Warning: This will destroy the deployment!
File: agents.yaml
Service: praisonai-app

Type 'yes' to confirm destruction: yes

🗑️ Destroying deployment...

✅ Deployment destroyed successfully

Deleted resources:
  • container: praisonai-app-abc123
  • image: praisonai-app:latest (local)

Environment Variables

Variable	Required	Description
OPENAI_API_KEY	Yes*	OpenAI API key
ANTHROPIC_API_KEY	No	Anthropic API key
GROQ_API_KEY	No	Groq API key

*Required if using OpenAI models.

Troubleshooting

Issue	Fix
Docker not running	Start Docker daemon
Build failed	Check Dockerfile syntax, run praisonai deploy doctor
Push failed	Check registry credentials
Port in use	Change port in agents.yaml
Container exits	Check logs: docker logs <container>

Manual Docker Commands (Optional)

These commands are for manual validation only. Use praisonai deploy for deployment.

# Build manually
docker build -t praisonai-app:latest .

# Run manually
docker run -p 8005:8005 -e OPENAI_API_KEY=$OPENAI_API_KEY praisonai-app:latest

# Check logs
docker logs <container_id>

# Stop
docker stop <container_id>

Related

• Deploy CLI Overview - All deploy commands
• API Deploy - Deploy as local API server
• AWS Deploy - Deploy to AWS ECS/Fargate
• GCP Deploy - Deploy to Google Cloud Run