Trace CrewAI with BotGauge

Records CrewAI runs, plus the LLM calls CrewAI makes through the OpenAI SDK.


Extra	`[crewai]`
Verified framework versions	crewai 1.15.22 and 1.15.28
Instrumentor installed by the extra	`openinference-instrumentation-crewai 1.1.21`
Zero-code run	Traced

Also verified: crewai 1.15.22 with instrumentor 1.1.18. The extra no longer installs that older instrumentor. See supported frameworks for how verification works.

In code

Call tracing.init(), attach the framework instrumentor to the provider it returns, and wrap the work in tracing.turn() and tracing.agent().

CrewAI calls its model through the OpenAI SDK, which the CrewAI instrumentor doesn’t trace. The [crewai] extra therefore also installs the OpenAI producer, and the install line pins the tested openai. CrewAI 1.15 needs Python 3.10 to 3.13 and doesn’t support 3.14. Set CREWAI_DISABLE_TELEMETRY=true to stop CrewAI’s own usage telemetry.

Install

terminal

pip install "botgauge[crewai]" crewai==1.15.28 openai==2.54.0

Add the code

app.py

import os

from botgauge import tracing
from crewai import LLM, Agent, Crew, Task
from openinference.instrumentation.crewai import CrewAIInstrumentor
from openinference.instrumentation.openai import OpenAIInstrumentor

provider = tracing.init()
CrewAIInstrumentor().instrument(tracer_provider=provider)
OpenAIInstrumentor().instrument(tracer_provider=provider)

llm = LLM(
    model=os.environ["OPENAI_MODEL"],
    base_url=os.environ["OPENAI_BASE_URL"],
    api_key=os.environ["OPENAI_API_KEY"],
)
agent = Agent(role="Assistant", goal="Answer briefly", backstory="You answer in one sentence.",
              llm=llm)
task = Task(description="In one sentence, what is OpenTelemetry?", expected_output="One sentence",
            agent=agent)

with tracing.turn(session_id="ses_crewai"), tracing.agent("crewai-agent"):
    result = Crew(agents=[agent], tasks=[task]).kickoff()
print(result)

tracing.shutdown()

Run

terminal

python app.py

Needs BOTGAUGE_API_KEY, OPENAI_MODEL, OPENAI_BASE_URL, OPENAI_API_KEY.

Zero-code

Leave your app unchanged and start it with botgauge instrument. The sample still wraps the work in tracing.turn() and tracing.agent() so spans carry session and agent ids.

Install

terminal

pip install "botgauge[crewai]" crewai==1.15.28 openai==2.54.0

Add the code

app.py

import os

from botgauge import tracing
from crewai import LLM, Agent, Crew, Task

llm = LLM(
    model=os.environ["OPENAI_MODEL"],
    base_url=os.environ["OPENAI_BASE_URL"],
    api_key=os.environ["OPENAI_API_KEY"],
)
agent = Agent(role="Assistant", goal="Answer briefly", backstory="You answer in one sentence.",
              llm=llm)
task = Task(description="In one sentence, what is OpenTelemetry?", expected_output="One sentence",
            agent=agent)

with tracing.turn(session_id="ses_crewai"), tracing.agent("crewai-agent"):
    result = Crew(agents=[agent], tasks=[task]).kickoff()
print(result)

Run

terminal

botgauge instrument python app.py

Needs BOTGAUGE_API_KEY, OPENAI_MODEL, OPENAI_BASE_URL, OPENAI_API_KEY.

View CrewAI spans live

Watch your CrewAI runs and underlying LLM calls appear in BotGauge.

Try BotGauge →

See versions not verified if you run a different version, capture and redaction to control exported content, and troubleshooting if spans are missing.

Frameworks

More

Trace CrewAI with BotGauge

In code

Zero-code

​In code

​Zero-code

​Related

In code

Zero-code

Related