{"version":"2.1.318","anchor":"endconversation-tool-reworked-for-inner-calls","canonical_anchor":"endconversation-tool-reworked-for-inner-calls","heading":"Ending a conversation from a nested call now ends only the issuing turn","tier":"internal","area":"REPL Mode","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.318\/e\/endconversation-tool-reworked-for-inner-calls","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.318","markdown":"### Ending a conversation from a nested call now ends only the issuing turn\n\nWhen the end_conversation tool is called from inside another call, Claude Code can now check which message sent it and end that turn\n\n**Unclear.** It is unclear what switches the new behaviour on, so whether it applies to a given user is not known.\n\n**What**\n\nThe `end_conversation` tool lets Claude stop a conversation. When this tool is called as an inner call, made from inside another call rather than directly, Claude Code now keeps track of which message issued it. It checks that message and ends the current turn. In other cases it stops the conversation the same way as before. This behaviour sits behind an internal feature check.\n\nBefore, there was only one way out: the whole conversation was stopped.\n\n**Why**\n\nThis affects how a conversation ends when the request to end it comes from a nested agent (a helper copy of Claude running inside the main one).\n\n- Area: REPL Mode\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}