The filtered trace view preserves the initiating agent while investigators need the failing one.
Existing trace continuity protects familiar filtering, while failed handoffs require clear responsibility at the downstream call.
Who you’d be doing this for
“I spent half an hour in the parent agent's trace, then found the failure was two calls downstream.”
Santiago Delgado · AI Application Engineer
Operates a production multi-agent application and investigates failed tool calls through traces.
What is at stake
Engineers investigating failed handoffs are filtering into the initiating agent, not the agent that made the failing tool call. You have to weigh a clearer ownership signal against preserving the trace filtering customers already use.
Why it isn’t already fixed
Every obvious fix costs something else. That’s the part you’d have to decide.
- familiar filtering vs clear handoff ownership
- trace continuity vs investigator speed
- narrow release scope vs complete investigation context
Why Datadog
Agent Observability connects agent traces with production telemetry, so unclear handoff ownership slows the debugging workflow it is meant to support.
Written with these in mind
Not your kind of problem? 40 more at Datadog, or browse every organization.
This is the setup. The work is inside.
Running it puts you in the room: the full situation and its constraints, stakeholders who push back in their own words, and the decisions that are yours to make. What you produce becomes a Day One Plan — work you can show someone instead of describing.