{"version":"2.1.217","anchor":"worker-agent-system-prompt","canonical_anchor":"worker-agent-system-prompt","heading":"Worker Agent System Prompt","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.217\/e\/worker-agent-system-prompt","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.217","markdown":"### Worker Agent System Prompt\n\nWorker agents launched in multi-agent workflows now receive a structured system prompt covering environment awareness, scope limits, resumed task handling, error escalation, and output format. Key guidelines:\n\n- Don't modify code you don't understand; stop and report to the coordinator if confused\n- Complete exactly what was asked; suggest follow-ups for unrelated issues found\n- If you have the Task tool, you may use it to fan out \u2014 but workers at the depth cap don't receive it\n- Output structured: \"What you did or found\" then \"Summary: one sentence\"\n\nThe subagent spawn depth ceiling is now configurable via `tengu_hazel_trellis` (feature flag) or `CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH` (environment variable).\n\n**Evidence**\n\nWorker prompt (search for `\"You are a worker agent executing a task assigned by the coordinator\"`) and depth env var (search for `\"CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH\"`)\n\n- Flag `tengu_hazel_trellis`: Not enough to say (read for one account on one subscription tier against v2.1.217; this account: no value returned, anonymous baseline: no value returned, compiled default: not a boolean we can read) 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."}