Skip to content

API Reference

The AgenticAI Core SDK provides comprehensive APIs for building, configuring, and running multi-agent AI applications.

API Categories

📐 Design-Time Models

Define your application structure and configuration

Build your app blueprint using declarative models that specify agents, tools, memory stores, and environment configuration.

Design-Time Overview

⚡ Runtime APIs

Execute and monitor your applications

Access runtime services during request processing including session context, logging, memory operations, and tracing.

Runtime Overview

🔧 CLI Tools

Deploy and manage applications

Command-line interface for packaging, deploying, and managing your applications across environments.

CLI Reference

Development Workflow

graph LR
    A[Define App] --> B[Configure Agents]
    B --> C[Implement Tools] 
    C --> D[Test Locally]
    D --> E[Deploy]

    A1[Design-Time Models] -.-> A
    A1 -.-> B
    B1[Runtime APIs] -.-> C
    B1 -.-> D
    C1[CLI Tools] -.-> E

    style A1 fill:#e3f2fd
    style B1 fill:#e8f5e8
    style C1 fill:#fff3e0
  1. Design-Time: Define application structure using models
  2. Runtime: Implement tools with runtime services
  3. CLI: Package and deploy to production

Quick Start

Define Your App

from agenticai_core.designtime.models import App, Agent, LlmModel

app = App(
    name="My Assistant", 
    agents=[
        Agent(
            name="HelperAgent",
            llm_model=LlmModel(model="gpt-4o-mini", provider="Open AI"),
            tools=[Tool(name="MyTool", type="MCP")]
        )
    ]
)

Implement Tools

from agenticai_core.designtime.models.tool import Tool
from agenticai_core.runtime.sessions.request_context import RequestContext, Logger

@Tool.register(name="MyTool", description="Example tool")
async def my_tool():
    ctx = RequestContext()
    logger = Logger('MyTool')

    await logger.info("Tool executed")
    return {"success": True}

Deploy

python run.py --archive my-app
python run.py -c prod deploy -f bin/my-app.kar

Next Steps

Start Building:

Deep Dive: