{"version":"2.1.292","anchor":"prompt-cache-break-telemetry-adds-resume-and-subagent-fields","canonical_anchor":"prompt-cache-break-telemetry-adds-resume-and-subagent-fields","heading":"Prompt-cache break reports now say whether a session was resumed or a subagent","tier":"internal","area":"Telemetry","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.292\/e\/prompt-cache-break-telemetry-adds-resume-and-subagent-fields","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.292","markdown":"### Prompt-cache break reports now say whether a session was resumed or a subagent\n\nThe usage report sent when the prompt cache breaks now records whether it came from a subagent or a resumed session, plus an outline of the messages\n\n**What**\n\nWhen the prompt cache breaks, Claude Code sends a usage report about it. A cache break means the model has to reprocess parts of the conversation it would otherwise have reused. That report now also records:\n\n- whether the request came from a subagent, a helper agent that Claude Code starts to work on part of a task\n\n- whether the session was resumed, meaning an earlier conversation was reopened\n\n- whether that resumed session was reopened in a new process, meaning Claude Code had been closed and started again\n\n- an outline of the messages involved\n\nThe way Claude Code works out what broke the cache has also been reworked for sessions whose earlier state was restored from disk.\n\n**Why**\n\nThis only changes what Claude Code reports about itself. It does not change how Claude Code behaves or what you see.\n\n- Area: Telemetry\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}