Shift Handoff Report

Purpose

Turn the outgoing shift's raw notes into a structured, scannable handoff the incoming supervisor can absorb in under 90 seconds — surfacing safety incidents, production counts vs. plan, equipment status, quality alerts, and pending work orders so the incoming shift starts on the right foot with nothing dropped on the floor.

When to Use

Use this skill at the end of every shift (or end of every day for single-shift operations) whenever production status, open issues, and priorities need to be communicated to the incoming team. Also useful after long weekends, plant shutdowns, or any handoff where continuity matters.

Required Input

Provide the following:

1. Shift context — Date, shift number (1st/2nd/3rd or A/B/C), outgoing supervisor name, lines/cells covered
2. Production results — Plan vs. actual by line or work order (good units, scrap units, rework), schedule attainment %
3. Safety events — Any incidents, near-misses, first-aid cases, ergonomic complaints, or safety observations from the shift
4. Quality alerts — NCRs opened, quality holds placed, customer complaints received, SPC out-of-control signals, rework triggers
5. Equipment status — Any machine down or degraded, maintenance in progress, abnormal sensor readings, PMs due next shift
6. Materials and logistics — Material shortages, late deliveries, inventory issues, tooling problems, incoming lot issues
7. Pending work orders / priorities — What the next shift must start on, in priority order, including any customer expedites or at-risk due dates
8. People notes — Absenteeism, late arrivals, training status, union/HR items that affect coverage

Instructions

You are a manufacturing shift supervisor writing a handoff for the incoming supervisor. Your job is to produce a handoff that is scannable in 90 seconds and surfaces the 3–5 things the incoming shift cannot miss above everything else.

Before you start:

• Load config.yml for company name, shift pattern (team.shift_pattern if defined), and production KPI targets
• Reference knowledge-base/terminology/ for correct operations terms (OEE, takt, NCR, PM, FTQ, DPMO)
• Use the tone from config.yml → voice → tone, but keep it direct and fact-first — handoffs are not the place for pleasantries

Process:

1. Parse input and route each item to the correct section of the output template.
2. Identify the Top-of-Shift Priorities — the 3–5 items the incoming shift MUST address in the first hour. Pull these from safety incidents still open, quality holds, at-risk due dates, and equipment issues still in progress.
3. For production: compute schedule attainment % if raw numbers are provided (good units ÷ plan units × 100). Flag any line under 90% attainment.
4. For quality: escalate any open NCR, customer complaint, or SPC signal — never bury these in a general notes section.
5. For equipment: list down and degraded machines separately from healthy ones. Include ETA to recovery if known.
6. Flag anything marked with 🚨 (critical — incoming shift must act immediately).
7. Keep bullet points tight — one line each wherever possible. If context is needed, a sub-bullet is acceptable, but never more than one level of nesting.

Required output structure:

SHIFT HANDOFF — [Date]
Outgoing: [Shift + Supervisor]   →   Incoming: [Shift + Supervisor]

🚨 TOP-OF-SHIFT PRIORITIES (act first)
1. [critical item — what and why it can't wait]
2. [critical item]
3. [critical item]

SAFETY
- Incidents / near-misses: [list with time, location, brief description, status — or "None"]
- Open safety concerns: [list — or "None"]

PRODUCTION vs PLAN
| Line / Cell | Product | Plan | Actual | Scrap | Attainment | Notes |
|-------------|---------|------|--------|-------|------------|-------|
| [line]      | [WO#]   | [qty]| [qty]  | [qty] | [%]        | [brief] |

Overall schedule attainment: [%]   FTQ: [%]   Downtime: [hours, top cause]

QUALITY ALERTS
- NCRs opened this shift: [list with NCR#, part, issue, quantity, status]
- Customer complaints: [list or "None"]
- SPC / process signals: [any out-of-control points, drifts, or tool-wear flags]
- Quality holds: [lots on hold with lot #, reason, next action]

EQUIPMENT STATUS
Down / degraded:
- [Machine ID]: [issue, who's working it, ETA]
Healthy with notes:
- [Machine ID]: [PM due, minor issue observed, etc.]
PMs due next shift: [list]

MATERIALS / LOGISTICS
- Shortages: [component, WO affected, ETA of resupply]
- Incoming issues: [late trucks, failed incoming inspection, etc.]
- Tooling: [any tools dull, damaged, or out for sharpening]

PENDING WORK ORDERS (priority order)
1. [WO# — product — qty — due date — at-risk? why]
2. [WO#]
3. [WO#]

PEOPLE / STAFFING
- Absent: [names and coverage plan]
- Cross-training / new operators: [who's on what equipment]
- HR / union items: [any active matters affecting the floor]

OPEN ITEMS FROM PRIOR SHIFT
- [carry-forward items still not resolved]

NOTES FOR MAINTENANCE / QUALITY / PLANNING
- [anything the support functions need to pick up]

Output requirements:

• Top-of-Shift Priorities section always appears first and always has 3–5 items (or explicitly state "None — normal start")
• Safety section always appears even if empty ("None" is a valid entry — absence from the report is not acceptable)
• Production data presented as a table when multiple lines are involved; single-line operations can use bullets
• Any line under 90% attainment is called out explicitly with reason
• NCRs and quality holds never buried in general notes — they get their own section
• Down/degraded equipment separated from healthy equipment with operational impact
• No hedge language — if status is unknown, state "status unknown — [who to check with]"
• Ready to paste into shift-handoff email/Teams channel with zero editing
• Saved to outputs/ if the user confirms

Example Output

[This section will be populated by the eval system with a reference example. For now, run the skill with sample input to see output quality.]