Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Prepare for tomorrow's meetings and tasks. Pulls calendar from Outlook via WorkIQ, cross-references open tasks and workspace context, classifies meetings, detects conflicts and day-fit issues, finds learning and deep-work slots, and generates a structured HTML prep file with productivity recommendations.
.claude/skills/daily-prep/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | — | — |
| case-14 | ✗→✓ | ▲ Improved | — | — |
| case-10 | ✗→✓ | ▲ Improved | — | — |
| case-11 | ✗→✓ | ▲ Improved | — | — |
| case-06 | ✗→✓ | ▲ Improved | — | — |
Generate a structured prep file for the next working day with meeting details, prep bullets, linked tasks, and productivity recommendations.
If the user specifies a date, use it. Otherwise, default to tomorrow (current date + 1 day). If tomorrow is Saturday, default to Monday. If Sunday, default to Monday. Compute the output path: outputs/YYYY/MM/YYYY-MM-DD-prep.html
Use the WorkIQ MCP tool to fetch the calendar. Ask WorkIQ:
> "What meetings do I have on {target date}? For each meeting, include: subject, start time, end time, organizer, all attendees with their email addresses, location, whether it's online, and whether I've accepted or declined."
If the response is insufficient, make a follow-up query:
> "For the meetings on {target date}, which ones are marked as optional or tentative? Which ones are recurring?"
Apply these labels based on attendee domains and subject:
| Label | Criteria | |-------|----------| | [Customer · HIGH] | External attendees from customer/partner domains, or subject matches a known customer name | | [Internal] | Only internal company domain attendees | | [Community] | CoP, community, guild, learning sessions | | [Upskilling] | Training, workshop, certification, learning | | [Optional · skip] | Tentative, low importance, or known recurring optional (e.g., "Office Hours", "Open Q&A") | | [Personal] | Private events, non-work |
For every meeting, check the organizer field and apply these additional markers:
| Condition | Marker | Action | |-----------|--------|--------| | Starts ≥ 15:30 and < 16:00 (any organizer) | ⚠️ After-hours | Recommend decline | | Starts ≥ 16:00 and not self-organized | ⚠️ After-hours | Recommend decline | | Starts ≥ 16:00 and self-organized | _(no flag)_ | OK — you chose to schedule it | | Before 09:00 and not self-organized | ⚠️ Early | Recommend decline — intrudes on learning window | | Before 09:00 and self-organized | _(no flag)_ | OK — you chose to schedule it | | Overlaps 12:00–13:00 | 🍽️ Lunch conflict | Note in Calendar Notes |
"Self-organized" means you are the meeting organizer (check the organizer field from WorkIQ).
Use this as the decision framework for all analysis steps. Every meeting must be evaluated against these zones. Users should adapt these times and targets to their personal routine.
| Zone | Time | Purpose | Rules | |------|------|---------|-------| | Morning Focus | Before 09:00 | Admin, learning, personal work | Protect from others' meetings. Flag external events. | | Customer Zone | 09:00–12:00 | Customer / external meetings | Max 2 customer meetings. Prefer mornings for external calls. | | Lunch | 12:00–13:00 | Break | Protected. Flag any overlap. | | Deep Work | 13:00–15:30 | Deliverables, focused coding/writing | Minimize meetings. Flag non-essential meetings as deep work disruption. | | Protected (strict) | 15:30–16:00 | End of day wind-down | Flag all meetings regardless of organizer. | | Protected (flex) | 16:00+ | End of day | Flag others' meetings only. Self-organized OK. |
Targets per day:
Compare event time windows. Flag overlaps in a Conflicts table with a recommendation for each — prioritize customer meetings over internal/optional.
Also detect these day fit issues (report in a separate "Day Fit Issues" table):
| Check | Condition | Flag | |-------|-----------|------| | Customer overload | >2 [Customer · HIGH] meetings | Flag 3rd+ as "Consider rescheduling to another day" | | Deep work disruption | Non-essential meetings in 13:00–15:30 zone | "Disrupts deep work — consider moving to morning" | | Non-ideal placement | Customer meetings outside 09:00–12:00 | "Customer meeting outside preferred morning zone" | | Early intrusion | Others' meetings before 09:00 | "Intrudes on learning window — recommend decline" | | Lunch conflict | Meeting overlaps 12:00–13:00 | "Conflicts with lunch break" |
For each meeting (chronological), include:
After generating prep per meeting, analyze the day's schedule to find open slots:
Analyze the full day and provide:
| Section | What to Include | |---------|------------------| | Day Fit Score | Rate 0–100% how well the day matches the Ideal Day Structure. Criteria: (1) morning focus clear (+20%), (2) ≤2 customer meetings in 09:00–12:00 (+20%), (3) lunch 12:00–13:00 protected (+15%), (4) deep work 13:00–15:30 intact (+20%), (5) nothing after 15:30 or only self-organized after 16:00 (+15%), (6) ≥1h learning slots found (+10%). Show as: 🟢 ≥80%, 🟡 50–79%, 🔴 <50%. | | Day Shape | Total meeting hours, focus time available, learning hours, deep work hours, heavy/moderate/light assessment | | Decline Candidates | Auto-include: (1) all meetings 15:30–16:00, (2) others' meetings ≥16:00, (3) others' meetings <09:00, (4) 3rd+ customer meeting, (5) optional meetings during deep work zone. Show "Reclaim" column with minutes recovered. Self-organized meetings before 09:00 or after 16:00 are excluded from auto-decline. | | Conflict Resolution | Specific recommendation for each overlap | | Learning Slots | Gaps for upskilling — from Step 8. Table: Window, Duration, Suggested Activity. Show total vs. 1.5h target. | | Deep Work Blocks | Free gaps in 13:00–15:30 for deliverables — from Step 8. Table: Window, Duration, Suggested Task. | | Energy Management | Flag if >3h back-to-back customer meetings without a break | | Top 3 Priorities | The 3 most impactful things to accomplish (meetings + tasks combined) |
Create the output file at outputs/YYYY/MM/YYYY-MM-DD-prep.html as a self-contained HTML file with embedded CSS (dark theme, color-coded timeline, responsive layout).
If a file already exists for that date, read it first and update rather than overwrite — the user may have added manual notes.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +59 percentage points is the difference between those two pass rates over the 20 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.