Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks including CrewAI, Agno, OpenAI Agents SDK, Langchain, Autogen, AG2, and CamelAI
Python agentagentopsagents-sdkaianthropicautogencost-estimationcrewaievalsevaluation-metricsgroqlangchainllmmistralollamaopenaiopenai-agents
# # Anthropic Async Example
#
# Anthropic supports both sync and async! This is great because we can wait for functions to finish before we use them!
#
# In this example, we will make a program called "Titan Support Protocol." In this example, we will assign our mech a personality type and have a message generated based on our Titan's health (Which we randomly choose). We also send four generated UUIDs which are generated while the LLM runs
# First, we start by importing Agentops and Anthropic
# %pip install agentops
# %pip install anthropic
# Setup our generic default statements
from anthropic import Anthropic
import agentops
from dotenv import load_dotenv
import os
import random
import asyncio
import uuid
# And set our API keys.
load_dotenv()
os.environ["AGENTOPS_API_KEY"] = os.getenv("AGENTOPS_API_KEY", "your_api_key_here")
os.environ["ANTHROPIC_API_KEY"] = os.getenv("ANTHROPIC_API_KEY", "your_anthropic_api_key_here")
#
# Now let's set the client as Anthropic and open an agentops trace!
client = Anthropic()
agentops.init(trace_name="Anthropic Async Example", tags=["anthropic-async", "agentops-example"])
# Now we create three personality presets;
#
# Legion is a relentless and heavy-hitting Titan that embodies brute strength and defensive firepower, Northstar is a precise and agile sniper that excels in long-range combat and flight, while Ronin is a swift and aggressive melee specialist who thrives on close-quarters hit-and-run tactics.
TitanPersonality = [
"Legion is a relentless and heavy-hitting Titan that embodies brute strength and defensive firepower. He speaks bluntly.,",
"Northstar is a precise and agile sniper that excels in long-range combat and flight. He speaks with an edge of coolness to him",
"Ronin is a swift and aggressive melee specialist who thrives on close-quarters hit-and-run tactics. He talks like a Samurai might.",
]
# And our comabt log generator! We select from four health presets!
TitanHealth = [
"Fully functional",
"Slightly Damaged",
"Moderate Damage",
"Considerable Damage",
"Near Destruction",
]
# Now to the real core of this; making our message stream! We create this as a function we can call later! I create examples since the LLM's context size can handle it!
Personality = {random.choice(TitanPersonality)}
Health = {random.choice(TitanHealth)}
async def req():
# Start a streaming message request
stream = client.messages.create(
max_tokens=1024,
model="claude-3-7-sonnet-20250219",
messages=[
{
"role": "user",
"content": "You are a Titan; a mech from Titanfall 2. Based on your titan's personality and status, generate a message for your pilot. If Near Destruction, make an all caps death message such as AVENGE ME or UNTIL NEXT TIME.",
},
{
"role": "assistant",
"content": "Personality: Legion is a relentless and heavy-hitting Titan that embodies brute strength and defensive firepower. He speaks bluntly. Status: Considerable Damage",
},
{
"role": "assistant",
"content": "Heavy damage detected. Reinforcements would be appreciated, but I can still fight.",
},
{
"role": "user",
"content": "You are a Titan; a mech from Titanfall 2. Based on your titan's personality and status, generate a message for your pilot. If Near Destruction, make an all caps death message such as AVENGE ME or UNTIL NEXT TIME.",
},
{
"role": "assistant",
"content": f"Personality: {Personality}. Status: {Health}",
},
],
stream=True,
)
response = ""
for event in stream:
if event.type == "content_block_delta":
response += event.delta.text
elif event.type == "message_stop":
break # Exit the loop when the message completes
return response
async def generate_uuids():
uuids = [str(uuid.uuid4()) for _ in range(4)]
return uuids
# Now we wrap it all in a nice main function! Run this for the magic to happen! Go to your AgentOps dashboard and you should see this trace reflected!
async def main():
# Start both tasks concurrently
uuids, message = await asyncio.gather(generate_uuids(), req())
print("Personality:", Personality)
print("Health Status:", Health)
print("Combat log incoming from encrypted area")
print("Verification matrix activated.:")
for u in uuids:
print(u)
print(". Titan Message: ", message)
# Run the main function
asyncio.run(main())
# We can observe the trace in the AgentOps dashboard by going to the trace URL provided above.
# Let's check programmatically that spans were recorded in AgentOps
print("\n" + "=" * 50)
print("Now let's verify that our LLM calls were tracked properly...")
try:
agentops.validate_trace_spans(trace_context=None)
print("\n✅ Success! All LLM spans were properly recorded in AgentOps.")
except agentops.ValidationError as e:
print(f"\n❌ Error validating spans: {e}")
raise