Skip to content

API Reference

Welcome to the AgenticAI Core SDK API reference documentation. This section provides detailed information about all classes, methods, and functions available in the SDK.

Design-Time Models

Design-time models are used to define and configure your AI application structure.

Core Models

Model Description
App Top-level application configuration
Agent AI agent configuration and behavior
Tool Tool definitions for agent capabilities
LlmModel LLM provider and inference configuration
Prompt System and custom prompts
MemoryStore Persistent data storage configuration
AppConfiguration Advanced application features
AppNamespace Environment and context namespaces
AppVariable Application-level environment variables
Icon Visual identifiers for entities

Quick Navigation

Building Applications

from agenticai_core.designtime.models import App, Agent, AppNamespace, AppVariable

# Create an application with environment support
app = App(
    name="My App", 
    description="...",
    app_namespaces=[AppNamespace(name="production")],
    app_variables=[AppVariable(name="API_KEY", is_secured=True, value="$env.API_KEY", namespaces=["production"])]
)

App API Reference

Configuring Agents

from agenticai_core.designtime.models import Agent, LlmModel

# Create an agent
agent = Agent(
    name="MyAgent",
    llm_model=LlmModel(...),
    tools=[...]
)

Agent API Reference

Defining Tools

from agenticai_core.designtime.models.tool import Tool

# Register a custom tool
@Tool.register(name="my_tool", description="...")
def my_tool(param: str):
    return result

Tool API Reference

Environment Management

from agenticai_core.designtime.models import AppNamespace, AppVariable

# Create namespaces for different environments
prod_ns = AppNamespace(name="production", description="Production environment")
dev_ns = AppNamespace(name="development", description="Development environment")

# Create environment-scoped variables
api_key = AppVariable(
    name="API_KEY",
    is_secured=True,
    value="$env.API_KEY",
    namespaces=["production"]
)

AppNamespace API Reference AppVariable API Reference

Runtime APIs

Runtime APIs are used during application execution.

  • Agent Runtime - Execute agents and handle requests
  • Memory Manager - Access and manage memory stores
  • Request Context - Session and context management

Runtime Overview

Type Reference

All models use Pydantic for validation and serialization. Key types include:

  • StrictStr - String with strict validation
  • StrictInt - Integer with strict validation
  • StrictBool - Boolean with strict validation
  • StrictFloat - Float with strict validation

Serialization

All models support standard serialization methods:

# To dictionary
model_dict = model.to_dict()

# To JSON string
json_str = model.to_json()

# From dictionary
model = ModelClass.from_dict(data)

# From JSON string
model = ModelClass.from_json(json_str)

Builder Pattern

Many models provide builder classes for fluent configuration:

from agenticai_core.designtime.models import AppConfigBuilder, AppNamespace, AppVariable

app_dict = AppConfigBuilder() \
    .set_name("My App") \
    .set_description("Description") \
    .set_agents([agent1, agent2]) \
    .set_app_namespace(AppNamespace(name="production")) \
    .set_app_variable(AppVariable(name="API_KEY", is_secured=True, value="$env.API_KEY", namespaces=["production"])) \
    .build()

app = App(**app_dict)

Next Steps