All products

Plan your working day with AI

Choose a realistic set of tasks from your own notes, keep deferred choices visible and record the actual day without automatic monitoring

Plan your working day with AI around the hours you actually have. The clock-in skill reads your authorised task sources, checks dated commitments and asks about your priorities before presenting possible work. You choose what enters the plan.

Its useful output is a dated section in your existing task file, including time estimates, starting gates and reasons for deferring other work. Setup requires an authoritative list of active projects and explicitly configured file paths. At the end of the day, you report what happened; the agent does not infer completion from a file being open. A mid-day handover preserves the next step for another session without treating the working day as finished. There is no automatic calendar connection, reminder service or continuous monitoring.

What it does

  • Make task choices fit the time remaining after fixed commitments.
  • Define a starting gate as an observable artefact or time box.
  • Keep deferred tasks and their reasons beside the chosen work.
  • Record an end-of-day account in the existing local log.
  • Carry done, in-progress and next steps into a fresh session.
  • Treat optional file-activity clues as questions, never proof of attention.

Install and use

  1. Fill in the context template with authorised task files, dated commitments and the section that may be updated.
  2. For example, ask: “I have two hours before a meeting. Ask what matters today, then help me choose a small plan.”
  3. Choose the work yourself, agree rough durations and starting gates, and report actual outcomes when clocking out.

FAQ

Source

Made by Omar Musani. Original skill: clock-in, published on GitHub under the MIT licence.

View on GitHub