CrewAI Tracing & Multi-Agent Monitoring
Auto-instrument every CrewAI crew execution, agent by agent, task by task. Get per-agent span trees, tool call visibility, and full cost attribution across your multi-agent pipeline.
Free tier · No credit card · 2-minute setup
What you get with Fluiq for CrewAI
Per-agent span tree
Each agent's task execution is a separate span, see which agent consumed the most tokens, took the longest, or called the most tools in a single crew kickoff.
Tool call tracing
Every tool invocation by a CrewAI agent is a child span with the tool name, input arguments, output, and latency, full visibility into what your agents are actually doing.
Crew-level cost rollup
Total token and USD cost aggregated across the entire crew run, broken down by agent and task, know the cost of each kickoff before it hits your billing statement.
Add Fluiq to your CrewAI app in 2 lines
import fluiq
fluiq.instrument(api_key="fl_...") # patches CrewAI automatically
from crewai import Agent, Task, Crew
from crewai.tools import BaseTool
researcher = Agent(
role="Senior Researcher",
goal="Find the latest AI ops trends",
backstory="You are an expert researcher",
llm="gpt-4o",
)
task = Task(description="Research LLM observability tools", agent=researcher)
crew = Crew(agents=[researcher], tasks=[task], verbose=True)
# Per-agent spans with cost attribution, no code changes:
result = crew.kickoff()What Fluiq instruments in CrewAI
Every call to these methods is automatically traced, no decorators, no wrappers, no manual spans.
Crew.kickoff() / kickoff_async()Agent._execute_task()Task.execute() / execute_sync()BaseTool._run() / _arun()Agent.execute_task()CrewAgentExecutor._call()Memory.save() / search()Start tracing CrewAI in 2 minutes
Free tier. No credit card. Full traces, security scanning, and evals on your first CrewAI call.
50,000 free traces / month · 1,000 evals / month · 14-day retention