What Prefactor records from CrewAI
CrewAI + Prefactor
Observe for CrewAI
Prefactor observes your CrewAI agents in real time — every LLM call, tool invocation, and custom span captured as struct
Open →EvaluateEvaluate for CrewAI
Prefactor evaluates your CrewAI agents — score outcome quality against the captured spans, track drift by comparing cust
Open →ObserveAct for CrewAI
Prefactor acts on your CrewAI agents at runtime — block, throttle, sandbox, or escalate a tool call or data access befor
Open →How the CrewAI integration works
- Each Task and the Agent assigned to it is captured as a nested span, with delegations preserved in the timeline — including which Process a Crew used to order them: Sequential, or Hierarchical with a Manager Agent dynamically assigning work.
- For Crews built as Flows — the event-driven layer wrapping @start/@listen/@router methods — each step is its own span, so the same trace covers direct LLM calls and full Crew runs inside one Flow.
- Beyond auto-captured spans, use withSpan to record any custom step you define — an API call, a quality check, a business action.
CrewAI integration FAQ
Do I need a dedicated package for CrewAI?
You can instrument CrewAI today with the framework-agnostic prefactor-core SDK; a dedicated package can be added on request.
What does Prefactor capture from CrewAI?
Prefactor records agent runs, task executions, delegations between agents, tool calls and LLM calls as structured, timestamped spans — so every CrewAI run is captured as trace data you can reconstruct, search and export end to end.
Does Prefactor add latency or change how CrewAI runs?
No. Observability capture is designed to stay off your agent's critical path, so it doesn't alter your CrewAI logic or your users' responses. The only part that acts inline is the optional runtime guardrails you enable per agent — by design, so a high-risk or low-confidence action can be held for human approval before it executes.
Can I evaluate agents built with CrewAI and catch regressions?
Yes. Once runs are captured, eval suites score quality and groundedness on real traffic, drift detection flags behaviour changes after deployment, and versioned eval history catches regressions before they ship — the observe → evaluate → improve loop applied to your CrewAI agents.
Related guides
See it on your CrewAI agents
Book a 15-minute setup and our team gets you tracing in production.