Creating Agents¶

Learn how to create and configure AI agents in the AgenticAI Core SDK.

Overview¶

Agents are the core building blocks of your application. Each agent is an autonomous entity with: - Specific role and responsibilities - LLM configuration for decision-making - Tools to perform actions - Prompts that define behavior

Agent Types¶

Autonomous Agents¶

AI-powered agents with LLM decision-making:

from agenticai_core.designtime.models.agent import Agent
from agenticai_core.designtime.models.llm_model import LlmModel, LlmModelConfig
from agenticai_core.designtime.models.prompt import Prompt

agent = Agent(
    name="FinanceAssist",
    description="Banking assistant for account management",
    role="WORKER",
    sub_type="REACT",
    type="AUTONOMOUS",
    llm_model=LlmModel(
        model="gpt-4o",
        provider="Open AI",
        connection_name="Default Connection",
        modelConfig=LlmModelConfig(
            temperature=0.7,
            max_tokens=1600
        )
    ),
    prompt=Prompt(
        system="You are a helpful banking assistant.",
        custom="Assist with account inquiries and transactions."
    )
)

Proxy Agents¶

Delegate to external systems:

agent = Agent(
    name="ExternalService",
    description="Integrates with external API",
    role="WORKER",
    sub_type="PROXY",
    type="AUTONOMOUS",
    llm_model=...,
    prompt=...
)

Agent Sub-Types¶

REACT (ReAct Pattern)¶

Reasoning + Acting pattern where the agent iteratively: 1. Reasons about the task 2. Selects an action (tool) 3. Observes the result 4. Continues until complete

agent = Agent(
    name="ReactAgent",
    sub_type="REACT",
    ...
)

Best for: - General purpose tasks - Interactive problem-solving - Tool-heavy workflows

PLAN_EXECUTE¶

Creates a plan first, then executes steps:

agent = Agent(
    name="PlannerAgent",
    sub_type="PLAN_EXECUTE",
    ...
)

Best for: - Complex multi-step tasks - Structured workflows - Tasks requiring upfront planning

Agent Roles¶

WORKER¶

Executes specific tasks delegated by the orchestrator:

agent = Agent(
    name="BillingWorker",
    role="WORKER",
    description="Handles billing and payment tasks",
    ...
)

SUPERVISOR¶

Coordinates other agents (typically the orchestrator):

agent = Agent(
    name="Supervisor",
    role="SUPERVISOR",
    description="Routes requests to appropriate workers",
    ...
)

Adding Tools to Agents¶

Using Builder Pattern¶

from agenticai_core.designtime.models.agent import AgentConfigBuilder
from agenticai_core.designtime.models.tool import Tool

# Define tools
tool1 = Tool(name="GetBalance", type="toolLibrary", ...)
tool2 = Tool(name="Transfer", type="toolLibrary", ...)

# Build agent with tools
agent_config = AgentConfigBuilder() \
    .set_name("BankingAgent") \
    .set_description("Banking operations agent") \
    .set_role("WORKER") \
    .set_sub_type("REACT") \
    .set_type("AUTONOMOUS") \
    .set_llm_model(llm_model) \
    .set_prompt(prompt) \
    .set_tool(tool1) \
    .set_tool(tool2) \
    .build()

agent = Agent(**agent_config)

Direct Assignment¶

agent = Agent(
    name="BankingAgent",
    tools=[tool1, tool2, tool3],
    ...
)

Configuring Agent Prompts¶

Basic Prompt¶

prompt = Prompt(
    system="You are a helpful assistant.",
    custom="You help with banking operations."
)

With Instructions¶

prompt = Prompt(
    system="You are a banking assistant.",
    custom="""You help customers manage accounts and transactions.

    ## Your Capabilities
    - Check account balances
    - Process transactions
    - Answer policy questions
    """,
    instructions=[
        "Never ask for passwords or PINs",
        "Always confirm transaction amounts",
        "Be professional and courteous"
    ]
)

With Memory Context¶

prompt = Prompt(
    system="You are a helpful assistant.",
    custom="""You have access to customer information:

    <AccountInfo>
    {{memory.accountInfo.accounts}}
    </AccountInfo>

    Use this to provide quick responses without invoking tools.
    """
)

Agent Icon Configuration¶

from agenticai_core.designtime.models.icon import Icon

agent = Agent(
    name="FinanceAssist",
    icon=Icon(
        name="avatar3",
        color="#D9D6FE",
        type="human"
    ),
    ...
)

Real-Time Configuration¶

Enable voice/audio capabilities:

agent = Agent(
    name="VoiceAgent",
    is_real_time_enabled=True,
    ...
)

Agent Metadata¶

Add custom metadata:

agent = Agent(
    name="CustomAgent",
    metadata={
        "versionInfo": "Draft version",
        "customField": "value"
    },
    ...
)

Converting to AgentMeta¶

For orchestrator registration:

# Convert full Agent to lightweight AgentMeta
agent_meta = agent.to_agent_meta()

# Use in orchestrator
app_agents = [agent.to_agent_meta() for agent in agents]
orchestrator = CustomOrchestrator(agents=app_agents)

Best Practices¶

Naming
Use descriptive names that reflect the agent's purpose
Keep names concise but meaningful
Descriptions
Provide detailed capability descriptions
Help the orchestrator understand when to use the agent
Include limitations and constraints
LLM Configuration
Match temperature to task type
Set appropriate token limits
Consider cost vs. quality
Prompts
Be specific about the agent's role
Include relevant context
Provide clear guidelines
Use memory template variables
Tools
Only include relevant tools
Avoid tool overload
Ensure tools align with agent's purpose
Testing
Test agents individually
Verify tool integration
Check memory access
Validate prompt effectiveness

Send Feedback