Examples

import asyncio
from echo import (
    GenericAgent, AgentConfig, LLMConfig, PersonaConfig, TaskConfig,
    ConversationContext, Message, MessageRole, TextMessage
)

async def main():
    agent = GenericAgent(
        agent_config=AgentConfig(
            persona=PersonaConfig(
                role="Health Educator",
                goal="Explain health topics simply",
                backstory="A friendly health educator",
            ),
            task=TaskConfig(
                description="Answer health questions clearly",
                expected_output="Simple, accurate explanations",
            ),
        ),
        llm_config=LLMConfig(provider="openai", model="gpt-4o-mini"),
        tools=[],
    )

    context = ConversationContext()
    context.add_message(Message(
        role=MessageRole.USER,
        content=[TextMessage(text="What is cholesterol?")],
    ))

    result = await agent.run(context)
    print(result.llm_response.text)

asyncio.run(main())

import asyncio
from echo import get_llm, LLMConfig, StreamEventType, ConversationContext
from echo_agents.tools.elicitation import SelectionElicitationTool

async def main():
    llm = get_llm(LLMConfig(provider="openai", model="gpt-4o-mini"))
    tools = [SelectionElicitationTool()]
    context = ConversationContext()

    system_prompt = """You are a medical assistant collecting symptoms.
    Use elicit_selection to present symptom options as buttons."""

    context.add_message(Message(
        role=MessageRole.USER,
        content=[TextMessage(text="I'm not feeling well")],
    ))

    print("Assistant: ", end="", flush=True)

    async for event in llm.invoke_stream(context, tools=tools, system_prompt=system_prompt):
        if event.type == StreamEventType.TEXT:
            print(event.text, end="", flush=True)
        elif event.type == StreamEventType.TOOL_CALL_START:
            print(f"\n[Tool: {event.json.get('tool_name')}]")
        elif event.type == StreamEventType.DONE:
            print()
            if event.llm_response.elicitations:
                elicit = event.llm_response.elicitations[0]
                print(f"\n📋 {elicit.details.input.get('text')}")
                for opt in elicit.details.input.get('options', []):
                    print(f"   • {opt}")

asyncio.run(main())

import asyncio
import os
from echo import (
    GenericAgent, AgentConfig, LLMConfig, PersonaConfig, TaskConfig,
    MCPConnectionManager, MCPServerConfig, MCPTransport,
    ConversationContext, Message, MessageRole, TextMessage
)

async def main():
    # Connect to Eka MCP
    manager = MCPConnectionManager(MCPServerConfig(
        transport=MCPTransport.STREAMABLE_HTTP,
        url="https://mcp.eka.care/mcp",
        headers={"x-eka-jwt-payload": os.getenv("EKA_JWT_PAYLOAD")},
    ))

    mcp_tools = await manager.get_tools()
    print(f"Tools: {[t.name for t in mcp_tools]}")

    agent = GenericAgent(
        agent_config=AgentConfig(
            persona=PersonaConfig(
                role="Medical Assistant",
                goal="Help with patient queries using tools",
                backstory="A helpful medical AI with access to patient data",
            ),
            task=TaskConfig(
                description="Use tools to answer queries accurately",
                expected_output="Information from tools when appropriate",
            ),
        ),
        tools=mcp_tools,
        llm_config=LLMConfig(provider="openai", model="gpt-4o-mini", max_iterations=5),
    )

    context = ConversationContext()
    context.add_message(Message(
        role=MessageRole.USER,
        content=[TextMessage(text="Search for patient Kumar")],
    ))

    result = await agent.run(context)

    # Show tool calls
    for item in result.llm_response.verbose:
        if item.type == "tool":
            print(f"[Tool] {item.tool_name}")

    print(f"\nAssistant: {result.llm_response.text}")

    await MCPConnectionManager.cleanup_all()

asyncio.run(main())

async def conversation():
    agent = GenericAgent(
        agent_config=agent_config,
        llm_config=llm_config,
        tools=tools,
    )

    context = ConversationContext()

    for i in range(10):
        user_input = input(f"[{i+1}] You: ").strip()
        if user_input.lower() == "quit":
            break

        context.add_message(Message(
            role=MessageRole.USER,
            content=[TextMessage(text=user_input)],
        ))

        result = await agent.run(context)
        print(f"Assistant: {result.llm_response.text}\n")

        # Update context for next turn
        context = result.context