The session detail page is the full record of a single agent run. Click any row in the Sessions view to open it.
What you see
The detail page surfaces:
- Chain — the full sequence of user and assistant turns.
- Tool calls — every tool the agent invoked.
- Prompt and reference thread — the original prompt plus any reference thread the session forked from.
- Telemetry — model, tokens, cost, duration, and completion status.
- Author — the team member who ran the session.
Use the detail page to understand how a result was produced, not just what was produced.
Ways teams use this data
Captured sessions give your team something AI-generated code usually hides: how the work happened. Use that visibility to:
Make AI work transparentTurn opaque agent runs into a record your team can actually read. Useful when reviewing what shipped and when answering “how did we end up here?”
Reuse what workedWhen a teammate solves a hard problem, the prompt, tool order, and turns are right there. Skip the rediscovery.
Debug with full contextFailed runs show the chain that led to the dead end — context lost, wrong tool, missing step.
Spot expensive patternsSurface sessions that burned tokens, looped, or retried so you can cut waste before it scales across the team.
Onboard with real workShare real sessions instead of stale best-practice docs. New teammates see how senior teammates actually use the agent.
Build a team knowledge baseA growing history is the most accurate documentation of how your team works with AI — raw material for skills, prompts, and playbooks.