Explanation
The explanation section provides understanding-oriented documentation that clarifies concepts and design decisions in cyberian.
What are Explanations?
Explanations help you understand:
- How things work under the hood
- Why things are designed the way they are
- Connections between different components
- Background and context for features
If you need step-by-step instructions, see Tutorials or How-To Guides instead.
Available Topics
Architecture
Understand cyberian's system architecture:
- How cyberian wraps agentapi
- HTTP communication model
- Client-server relationship
- Process management
- Server farms architecture
Workflow Model
Deep dive into workflow execution:
- Recursive task tree structure
- Depth-first traversal algorithm
- Completion detection mechanism
- Loop execution model
- Template rendering pipeline
- Success criteria validation
Agent Lifecycle
Understand agent state management:
- Server lifecycle (start, run, stop)
- Agent memory and context
reusevsrefreshmodes- When to use each mode
- State isolation patterns
Design Decisions
Learn why cyberian works the way it does:
- Why YAML for workflows?
- Why Jinja2 for templates?
- Why synchronous polling?
- Why recursive task trees?
- Trade-offs and alternatives
Reading Guide
These explanations are best read:
- After completing the tutorials
- When you're curious about how things work
- Before making architectural decisions in your workflows
- To understand error messages and behavior
See Also
- Tutorials - Step-by-step learning
- How-To Guides - Task-oriented recipes
- Reference - Complete specifications