BeeAI Tracing (Python)

Phoenix provides seamless observability and tracing for BeeAI agents through the Python OpenInference instrumentation package.

Install#

pip install openinference-instrumentation-beeai beeai-framework

Setup#

Connect your application to Phoenix with the register function:

from phoenix.otel import register# configure the Phoenix tracertracer_provider = register(project_name="beeai-agent", # Default is 'default'auto_instrument=True # Auto-instrument your app based on installed OI dependencies)

Run BeeAI#

Sample agent built using BeeAI with automatic tracing:

import asynciofrom beeai_framework.agents.react import ReActAgentfrom beeai_framework.agents.types import AgentExecutionConfigfrom beeai_framework.backend.chat import ChatModelfrom beeai_framework.backend.types import ChatModelParametersfrom beeai_framework.memory import TokenMemoryfrom beeai_framework.tools.search.duckduckgo import DuckDuckGoSearchToolfrom beeai_framework.tools.search.wikipedia import WikipediaToolfrom beeai_framework.tools.tool import AnyToolfrom beeai_framework.tools.weather.openmeteo import OpenMeteoToolllm = ChatModel.from_name(    "ollama:granite3.1-dense:8b",    ChatModelParameters(temperature=0),)tools: list[AnyTool] = [    WikipediaTool(),    OpenMeteoTool(),    DuckDuckGoSearchTool(),]agent = ReActAgent(llm=llm, tools=tools, memory=TokenMemory(llm))prompt = "What's the current weather in Las Vegas?"async def main() -> None:    response = await agent.run(        prompt=prompt,        execution=AgentExecutionConfig(            max_retries_per_step=3, total_max_retries=10, max_iterations=20        ),    )    print("Agent 🤖 : ", response.result.text)asyncio.run(main())

Observe#

Phoenix provides visibility into your BeeAI agent operations by automatically tracing all interactions.

Resources#

OpenInference package for Python