{"version":"2.1.293","anchor":"agent-tool-enablement-check-refactored","canonical_anchor":"agent-tool-enablement-check-refactored","heading":"Checks for when Claude can start a subagent now go through one place","tier":"internal","area":"Subagents","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293\/e\/agent-tool-enablement-check-refactored","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.293","markdown":"### Checks for when Claude can start a subagent now go through one place\n\nThe conditions that decide whether Claude can start a subagent are now combined into one shared check that also sees allowed agent types\n\n**Unclear.** It is not clear whether this changes when subagents can be started or only reorganises the same checks.\n\n**What**\n\nA subagent is a helper agent Claude can start to handle part of a task. The conditions that decide whether that option is offered used to be written out in place: whether the tool is present, its name, and a limit on how deeply agents can be nested. They are now handled by one shared check, which also receives the current nesting depth and the list of allowed agent types.\n\n**Why**\n\nThis could change when starting a subagent is offered.\n\n- Area: Subagents\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}