How Blueprints Work
Blueprints optimize the agent's context to prevent AI hallucinations
What are Blueprints?
Blueprints are step-by-step instructions you can customize. They're composed of different step types that optimize the agent's context for each step.
Blueprints are a new layer that improves results for cloud infrastructure tasks. Without blueprints, AI-generated code often contains:
- Deprecated API versions
- Missing security configurations
- Incorrect resource relationships
- Non-existent Terraform providers or modules
With blueprints, you get guided workflows instead of free-form generation:
Without Blueprints:
"Create a Kubernetes cluster" → ❌ Hallucinated, untested code
With Blueprints:
"Create a Kubernetes cluster" → ✅ Follows GKE/EKS blueprint → Production-ready code
Each blueprint contains:
- Step-by-step instructions - The agent knows what to do at each stage
- Configurable defaults - You can customize to fit your needs
- Error prevention - Common mistakes are caught before generation
Blueprint Types
infra.new has blueprints for the following use cases:
🏗️ Infrastructure Blueprints
Kubernetes clusters, databases, networking, and storage with proper configuration
🔄 Migration Blueprints
Import existing AWS/GCP resources into Terraform without manual translation
🔧 Maintenance Blueprints
Autofix drift and policy violations to keep your infrastructure secure and compliant
Trust, But Verify
While blueprints significantly improve code quality, always review generated code before deploying to production.
Review the Blueprint Plan
Before generation starts, review the blueprint plan to understand what will be created and make adjustments.
Enable Security Scanning (Optional)
Turn on Trivy security scanning to catch misconfigurations and vulnerabilities in generated code.
Set Up CI/CD with Presubmit Checks
Export your code to GitHub or GitLab and configure CI/CD pipelines with automated testing and validation. Ask the agent to set this up for you with presubmit checks to catch issues before deployment.
Monitor with Drift Detection
Enable drift detection to track infrastructure changes over time and get alerted when resources drift from their expected state.
💡 Best practice: Use a GitOps workflow with automated checks and approvals for production deployments. The agent can help you set this up.
How Blueprints Work
Blueprints optimize the agent's context to prevent AI hallucinations
What are Blueprints?
Blueprints are step-by-step instructions you can customize. They're composed of different step types that optimize the agent's context for each step.
Blueprints are a new layer that improves results for cloud infrastructure tasks. Without blueprints, AI-generated code often contains:
- Deprecated API versions
- Missing security configurations
- Incorrect resource relationships
- Non-existent Terraform providers or modules
With blueprints, you get guided workflows instead of free-form generation:
Without Blueprints:
"Create a Kubernetes cluster" → ❌ Hallucinated, untested code
With Blueprints:
"Create a Kubernetes cluster" → ✅ Follows GKE/EKS blueprint → Production-ready code
Each blueprint contains:
- Step-by-step instructions - The agent knows what to do at each stage
- Configurable defaults - You can customize to fit your needs
- Error prevention - Common mistakes are caught before generation
Blueprint Types
infra.new has blueprints for the following use cases:
🏗️ Infrastructure Blueprints
Kubernetes clusters, databases, networking, and storage with proper configuration
🔄 Migration Blueprints
Import existing AWS/GCP resources into Terraform without manual translation
🔧 Maintenance Blueprints
Autofix drift and policy violations to keep your infrastructure secure and compliant
Trust, But Verify
While blueprints significantly improve code quality, always review generated code before deploying to production.
Review the Blueprint Plan
Before generation starts, review the blueprint plan to understand what will be created and make adjustments.
Enable Security Scanning (Optional)
Turn on Trivy security scanning to catch misconfigurations and vulnerabilities in generated code.
Set Up CI/CD with Presubmit Checks
Export your code to GitHub or GitLab and configure CI/CD pipelines with automated testing and validation. Ask the agent to set this up for you with presubmit checks to catch issues before deployment.
Monitor with Drift Detection
Enable drift detection to track infrastructure changes over time and get alerted when resources drift from their expected state.
💡 Best practice: Use a GitOps workflow with automated checks and approvals for production deployments. The agent can help you set this up.