{"version":"2.1.293","anchor":"agent-tool-prompt-gets-a-continueavailable-input-tengu-velv","canonical_anchor":"agent-tool-prompt-gets-a-continueavailable-input-tengu-velv","heading":"Agent tool text now knows whether a subagent can be continued","tier":"internal","area":"Subagents","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293\/e\/agent-tool-prompt-gets-a-continueavailable-input-tengu-velv","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293","markdown":"### Agent tool text now knows whether a subagent can be continued\n\nThe text for the tool that starts subagents now receives a flag, worked out from the available tools, saying whether continuing is available\n\n**Unclear.** What the Agent tool's text says differently when continuing is available is not known.\n\n**What**\n\nThe Agent tool is how Claude starts a subagent, a helper copy of Claude given part of a task. The text Claude Code builds for this tool now receives a value saying whether continuing is available. Claude Code works that value out from the list of tools the session has.\n\nThe setting for keeping tool results inside subagents still depends on a remote switch, and it is still off unless that switch is turned on remotely.\n\n**Why**\n\nThis points to a way of continuing an existing subagent, and the Agent tool's wording may differ depending on whether that option exists in a session.\n\n- Flag `tengu_velvety_sprout`: Not enough to say (read for one account on one subscription tier against v2.1.293; this account: no value returned, anonymous baseline: no value returned, compiled default: off) These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.\n- Area: Subagents\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}