How-to Guides

Welcome to the Kubiya SDK How-to Guides. This section provides practical instructions for building and working with Docker-based tools, teammates, workflows, and infrastructure.

Tool Development

Learn how to create powerful, reusable tools that can be consumed by AI agents and applications:

• Creating Docker-Based Tools: Leverage existing Docker images to build tools
• Tool Input/Output Schema: Define clear schemas for your tools
• Secrets Management: Handle credentials securely in your tools

Workflow Development

Master the art of chaining tools and teammates together for complex processes:

• Multi-Tool Workflows: Combine tools into powerful workflows
• Error Handling: Build robust workflows that handle failures gracefully
• Workflow Testing: Verify your workflows work as expected

Deployment

Deploy your tools and workflows to various infrastructure environments:

• Local Execution: Run tools on your local Docker environment
• Kubernetes Deployment: Scale your tools on Kubernetes
• CI/CD Integration: Integrate tools into your CI/CD pipeline

CLI Usage

Master the Kubiya CLI for building, testing, and managing tools:

• CLI Overview: Get started with the Kubiya CLI
• Managing Tools: Build and test tools with the CLI
• Managing Sources: Organize your tool sources

Integration with LLM Applications

Learn how to make your tools available to any LLM application:

• MCP Integration: Expose tools via the Model Context Protocol
• OpenAI Function Calling: Use tools with OpenAI's function calling
• Custom LLM Integration: Integrate with any LLM system

Teammate Development

Create intelligent AI agents that use your tools:

• Building Teammates: Create specialized AI agents
• Teammate Configuration: Configure teammates for specific tasks
• Teammate Deployment: Deploy teammates to production

Common Use Cases

Step-by-step guides for common scenarios:

• DevOps Automation: Automate infrastructure management
• Data Processing: Build data pipelines with tools
• Security Operations: Create security workflows
• Customer Support Automation: Build AI support agents