Energy Scheduling Skill

Hours are not interchangeable — the 9am hour and the 3pm hour hold different people — yet calendars treat them as identical slots, which is how peak hours get spent on email (because it arrived) and the hardest thinking gets scheduled into the slump (because the slot was free). Energy scheduling is matching, in two steps: map the real curve (a week of light self-observation beats a lifetime of assumptions — many people are wrong about their own peaks) and match the work to it — deep/creative work on peaks, mechanical admin on troughs (it fits there fine — that's the point), meetings on the shoulders. The calendar rebuild is where the map becomes money.

What This Skill Produces

• The observation protocol — the week-long, three-times-daily energy check (10 seconds each) that produces the real map
• The energy map — the personal curve: peaks, troughs, shoulders, and the day-of-week effects
• The matching table — the user's actual work-types assigned to their curve zones
• The rebuilt week — the calendar re-shaped: blocks on peaks ([deep-work-blocking](../deep-work-blocking/SKILL.md) placement input), corridors on shoulders ([context-switch-budget](../context-switch-budget/SKILL.md) alignment), admin in the troughs

Required Inputs

Ask for these if not provided:

• The assumed curve — when they think they're sharp (recorded before observing, because the delta between assumed and observed is the finding half the time)
• The work-type inventory — what the role actually contains: deep/creative, analytical, interactive, mechanical; matching needs the categories
• The fixed constraints — the immovable meetings, the team's hours, the school run; the rebuild works inside reality
• Chronotype honesty — the morning-person mythology pressures night owls into fake 6am peaks; the observation protocol is the antidote, and the skill takes its side

Framework: The Mapping Rules

1. Observe before believing: one week, three daily checks (mid-morning, mid-afternoon, late day): energy 1–5 + one word of context — ten seconds each. The context column catches the confounds (the 2pm trough that's actually the post-lunch-meeting crash, the Thursday peak that's really deadline adrenaline) — the map wants the baseline curve, not the artifact of one week's chaos.
2. The map has four zones, personally shaped: peak (the 2–3 genuinely sharp hours — most people get one solid peak, morning or evening by chronotype) · shoulders (functional, social, fine) · trough (the slump — universal, personally timed) · and the day-of-week layer (Monday ramp, Friday fade — real for most). Two peaks a day is rare; planning as if you have them is the fantasy the observation kills.
3. Match by cognitive price: peaks → the work only you can do at full sharpness (creating, hard analysis, the [decision-meeting-format](../decision-meeting-format/SKILL.md)-grade decisions) · shoulders → meetings and collaboration (interaction generates its own energy — it's shoulder-priced) · troughs → the mechanical backlog (expense reports, [email-triage-system](../email-triage-system/SKILL.md) windows, filing) — which fits the trough honestly; the slump does admin fine, and knowing that removes both the guilt and the waste.
4. Rebuild the calendar to the map: the peak hours get blocked first ([deep-work-blocking](../deep-work-blocking/SKILL.md) takes placement from here), the meeting corridors land on shoulders (the counter-offer default aims there), the trough gets the standing admin window — and the one-week trial precedes evangelism, because the rebuild proves itself in output or it doesn't.
5. Defend the peak like the asset it is: the peak hours are the scarcest resource in the week — the [meeting-cost-meter](../meeting-cost-meter/SKILL.md) can price an hour, but a peak hour spent on email is the expensive version of the waste. The standing question for anything landing on a peak: "does this need me sharp, or just present?" — present-only work moves to a shoulder, by default.

Output Format

Energy Schedule: [name]

The Observation Week

[The 3× daily check format · the context column · assumed-curve recorded first: (their guess)]

The Map

[Peaks: (hours) · shoulders · trough · day-of-week layer · the assumed-vs-observed delta noted]

The Matching Table

The Rebuilt Week

[Peaks → blocks (placed) · shoulders → corridors · trough → the admin window · the one-week trial + output check]

Quality Checks

• The assumed curve was recorded before observation started
• The map cites a real observed week with confounds noted
• Every work-type has a zone with a cognitive-price reason
• Peak hours were blocked before anything else claimed them
• The trial week has an output check, not just a vibe check

Anti-Patterns

• Do not schedule by the assumed curve — half of assumed peaks are mythology, and the observation is cheap
• Do not spend peaks on email because it's there — arrival order is not a priority system
• Do not fight the trough with hard work — match it with the mechanical; the slump does admin fine
• Do not fake a chronotype — the 5am-club cosplay wastes a real evening peak
• Do not treat the map as permanent — re-observe when life shifts (new role, new kid, new season); curves move