← ClaudeAtlas

delegatelisted

Decides what to hand to an AI agent and what to keep, for one specific piece of work. Trigger when the user says "should I delegate this", "can AI do this", "should I automate this", "what should I hand off", "give this to an agent", "is this worth automating", "who should do this", "can Claude do this part", or describes a task and asks whether to run it themselves. Also trigger when a user is about to hand over work whose output they would not be able to check. Splits the work into parts, allocates each to the user, another person or a model, and attaches a verification step and a stop condition to every part that leaves the user's hands. Ends on a filled allocation table, not on advice. Use a process or SOP skill for documenting a repeatable procedure, and a ritual or automation-charter skill for a recurring team ceremony. This one is for a single piece of work.
theAnirudhKumar/work-design · ★ 0 · AI & Automation · score 70
Install: claude install-skill theAnirudhKumar/work-design
# Delegate Most delegation advice sorts tasks by whether a model can do them. That is the wrong sort. A model can do plenty of things whose output nobody in the room is able to check, and those are the ones that come back to bite. The failure this exists to prevent: **handing over the part you cannot verify, and finding out at delivery.** This skill takes one piece of work, splits it, and decides where each part goes. Every part that leaves the user's hands leaves with a check attached. It ends on a filled table. --- ## What this needs **Minimum: a description of the piece of work.** One or two sentences is enough. No connector, no workspace, no file access. **Better with** what the output is for and who sees it. The verification bar moves with the consequence, not with the task, so this changes the answer more than anything else the user could add. **Best with** one line on what the user has handed over before and what came back wrong. That turns generic checks into the specific ones that catch this user's actual failures. Where the user gives only the task and nothing else, run anyway and ask the three questions in Step 1 as part of the output rather than blocking on them. --- ## When this runs Runs when there is a specific piece of work on the table and the question is how to run it. **Does not run for:** - A recurring team ritual or a standing ceremony. That is a ritual inventory, and an automation-charter skill handles it. - Documenting a repeatable procedu