Solutions
Where each AI work handoff question is answered
Wagglet moves a prepared AI task to a qualified teammate who runs it on their own agent account. These pages take that one idea apart: what the category is, what happens when someone runs out of capacity, what survives a change of tool, how it sits beside a tracker you already use, and how an agent connects.
Start here
The category, then the four questions it raises
Read the first one if the idea is new. The other four answer a specific question and can be read in any order.
The categoryAI work handoffWhat an AI work handoff is, the five stages it moves through, and what stays with its owner when the task changes hands.Read the definition →CapacityShared AI execution capacityWhat to do when one person reaches their Claude Code or Codex limit and the work is stranded — and what Wagglet deliberately does not do.See how the task moves →Cross-tool contextClaude Code, Codex, and CursorWhat a task carries when it passes to a new session, a different agent, or a different person — and the one field it does not carry.See what has to move →Alongside your trackerWagglet and LinearLinear delegates an issue to an agent. Wagglet hands a prepared task to a person. When to run both, and when Linear alone is simpler.Compare the workflows →Agent accessMCP for coding agentsA bounded credential for one claimed task, or a named workspace connection for ongoing work — and what each one can reach.Choose the access model →