Observability Examples

This section demonstrates how to use observability tools with Kibo AI.

Langfuse Integration

Location: examples/observability/langfuse_example.py

This example shows how to integrate Langfuse for observability, tracing, and monitoring of your AI agents. It demonstrates how to initialize the Langfuse client and use it to track agent interactions.

import os
import sys

from dotenv import load_dotenv

sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), "../../src")))

from kiboai import AgentConfig, LangfuseConfig, create_agent
from kiboai.shared_kernel.logging import logger


def main():
    load_dotenv()
    logger.info("--- Kibo Langfuse Example ---")

    agent_def = AgentConfig(
        name="LangfuseBot",
        description="A helpful assistant with tracing.",
        instructions="Answer short and concise.",
        agent="agno",
        model="gpt-4o-mini",
        langfuse=LangfuseConfig(
            enabled=True,
            host=os.getenv("LANGFUSE_HOST", "http://localhost:3000"),
            public_key=os.getenv("LANGFUSE_PUBLIC_KEY"),
            secret_key=os.getenv("LANGFUSE_SECRET_KEY"),
            trace_name="kibo.langfuse_example",
        ),
    )

    api_key = os.getenv("OPENAI_API_KEY")
    if not api_key:
        logger.error("OPENAI_API_KEY is required to run this example.")
        return

    agent = create_agent(agent_def, api_key=api_key)
    result = agent.run("Hello from Kibo with Langfuse")

    logger.info("Response: %s", result.output_data)
    logger.info("Metadata: %s", result.metadata)


if __name__ == "__main__":
    main()

uv run examples/observability/langfuse_example.py