Handover field guide
Build AI work that another actor can continue
Practical, source-linked guides for agent handoffs, shared context, MCP workflows, and human review of AI-generated work.
Agent handoff
How to hand off context between AI agents
Use a durable continuation record to move objectives, state, decisions, evidence, and next actions between AI agents without replaying an entire chat.
Read guideContext engineering
Shared context vs agent memory
Understand when an agent needs private runtime memory and when a team needs an explicit, versioned context record shared across people and tools.
Read guideHuman review
How humans can review AI agent output
Anchor feedback to immutable files, assign clear follow-up work, and let agents retrieve and resolve review comments in the next revision.
Read guide