{"version":"2.1.66","anchor":"sub-agent-context-forking-forkcontext","canonical_anchor":"sub-agent-context-forking-forkcontext","heading":"Sub-Agent Context Forking ( forkContext )","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.66\/e\/sub-agent-context-forking-forkcontext","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.66","markdown":"### Sub-Agent Context Forking (`forkContext`)\n\n**What**\n\nAgents can now receive the full parent conversation context when spawned as sub-agents, enabling better awareness of what the user has been working on.\n\n**Details**\n\n- New `forkContext` frontmatter field for agent `.md` definitions (set to `true` or `false`)\n- When enabled, the sub-agent receives prior conversation messages as read-only context\n- Agents with `forkContext: true` must use `model: inherit` to avoid context length mismatch\n- A new `### FORKING CONVERSATION CONTEXT ###` boundary marker is injected to clearly separate inherited context from the sub-agent's own task\n- The sub-agent is instructed to only use tools available in its own system prompt, even if context messages reference other tools\n\n**Evidence**\n\nSub-agent entry system (search for `\"sub_agent_entered\"` and `\"FORKING CONVERSATION CONTEXT\"`)"}