{"version":"2.1.293","anchor":"agent-subagent-report-and-messaging-prompt-strings-refactore","canonical_anchor":"agent-subagent-report-and-messaging-prompt-strings-refactore","heading":"Subagent results can say whether the caller can send a follow-up","tier":"internal","area":"Subagents","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293\/e\/agent-subagent-report-and-messaging-prompt-strings-refactore","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293","markdown":"### Subagent results can say whether the caller can send a follow-up\n\nA subagent's result can now note that the calling agent cannot continue it, because the caller has no tool for sending it messages\n\n**What**\n\nA subagent is a helper agent that Claude Code starts to do part of a job and report back. Its result can now carry an internal field that is false when the agent that started it cannot continue it with a follow-up message, because that agent has no messaging tool. When the field is absent, a follow-up is possible.\n\nThe message shown when a subagent ends without delivering a report is now assembled from shared text instead of naming the `SendMessage` tool directly. A short description, \"Name for the spawned agent.\", was also added.\n\n**Why**\n\nThis affects how a subagent's result tells the agent that started it whether it can carry on the conversation with that subagent.\n\n- Area: Subagents\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}