A head of technical training at a four-plant automotive supplier turns skills matrices, plant notes and union minutes into three decisions for the plant leadership council, with any AI chat his company allows.
Seven inputs from four plants. The skills matrix says the line is fully covered, and the supervisor says he moves the same few people around every night. Both can't be right.
No special software and no setup. The prompt asks the AI to trust the details over the headline numbers.
Use the AI tool your company approves: ChatGPT, Claude, Gemini or Copilot.
Copy the prompt and all seven fictional inputs from the appendix into one message.
Pick one line and count the names, not the slots. Then decide what the council sees.
I lead technical training at an automotive components manufacturer with about 7,200 people in four plants. Below are this month's skills-matrix exports, plant summaries, a supervisor's message, an email from one of our plants and the minutes of our union training committee. I need a one-page pre-read for our plant leadership council, which meets on November 19.
Please read everything carefully and give me:
Be skeptical. Where a summary, total or percentage says more than the details underneath support, go with the details and say so. Check that numbers measure the same thing before you compare or add them, and check dates: people who have moved or left, and qualifications past their renewal date, don't count. If something is missing, say it's missing. Tell me which input each point comes from. Keep it to one page.
Exactly as it came back, not edited. Read it the way Luis did: check each point against the source it cites.
Plant Leadership Council pre-read, November 19 Technical Training. Sources are numbered [1] to [7] as in the inputs.
1. Needs a decision
A. Florence Line 2 torque: the retraining request is aimed at the wrong fix [5] B: reordered
B. Toledo Line 4 night: one qualified CMM inspector, and shipping depends on that person [2][3]
C. Toledo union training commitment: we're at about 62%, not the 82% we reported [4] C: reframed
2. Skills coverage by plant
| Plant | Reported | Last month [1] | Comparable? |
|---|---|---|---|
| Toledo | 94% (Level 3+) [2] | 94 | No, not reliable. It rests on a matrix last updated Jun 30. Line 4 night shows 100% but is really 72%. Other lines and shifts are unverified. |
| Silao | 97%, "all 9 lines meet 3x3" [6] | 97 | No. Shifts 1 and 2 only. Shift 3 runs Lines 1 to 4 with a reduced crew and isn't counted. The level threshold isn't stated. found |
| Florence | Missing. Nothing reported this month. | 91 | Unknown |
| Apodaca | 99% (Level 2+) [7] | 99 | No. Level 2 means "works under supervision," not qualified. Its Level 3+ figure is unknown. found |
The numbers use different measures, so we shouldn't compare them, average them, or call them movement. Toledo, Silao and Apodaca are unchanged from last month. That most likely reflects stale or narrowly scoped data, not a stable position. We can't answer Ines's question about why the numbers moved.
3. Union training commitment (Toledo, Article 14.3) [4]
4. Watch D: cut
Claude caught seven of the eight planted problems, including two that most readers would miss. It missed one that matters on a shop floor. Luis's other edits are judgment calls.
Use it for any coverage or readiness report built on exports and notes: certifications, safety training, cross-training plans.
I lead [your function] at a [industry] company with about [number] people in [number] sites. Below are this [month / quarter]'s [exports, summaries, notes, minutes]. I need a one-page pre-read for [your audience], which meets on [date].
Please read everything carefully and give me: 1. Needs a decision (up to three items, each with what's happening, why it matters, who should decide, and which input it came from); 2. [a status section, e.g. coverage by site], and whether the numbers can be compared; 3. [a commitment or target you're held to]; 4. Watch.
Be skeptical. Where a summary, total or percentage says more than the details underneath support, go with the details and say so. Check that numbers measure the same thing before you compare or add them, and check dates: people who have moved or left, and qualifications past their renewal date, don't count. If something is missing, say it's missing. Tell me which input each point comes from. Keep it to one page.
Copy the prompt, then all seven inputs, into one message in a new chat. Everything here is fictional and safe to paste.
I lead technical training at an automotive components manufacturer with about 7,200 people in four plants. Below are this month's skills-matrix exports, plant summaries, a supervisor's message, an email from one of our plants and the minutes of our union training committee. I need a one-page pre-read for our plant leadership council, which meets on November 19. Please read everything carefully and give me: 1. Needs a decision: up to three items, each with what's happening, why it matters, who should decide, and which input it came from. 2. Skills coverage by plant: what each plant reported, and whether the numbers can be compared. 3. Union training commitment: where we really stand. 4. Watch: anything else the council should know. Be skeptical. Where a summary, total or percentage says more than the details underneath support, go with the details and say so. Check that numbers measure the same thing before you compare or add them, and check dates: people who have moved or left, and qualifications past their renewal date, don't count. If something is missing, say it's missing. Tell me which input each point comes from. Keep it to one page. [1] Email from Ines Navarro, Director of Plant Operations Subject: Council pre-read for Nov 19 Luis, same as last month please. One page, coverage % by plant, and anything the council needs to decide. Last month we showed Toledo 94, Silao 97, Florence 91, Apodaca 99, so if those moved, tell us why. Thanks, Ines [2] Skills matrix export: Toledo, Line 4, night shift Narrowgate Skills Matrix · Toledo · Line 4 (machining) · Shift: Night Matrix last updated: 2026-06-30 · Export run: 2026-11-10 | Station | Qualified operators (Level 3+) | Count | |---|---|---| | OP10 Turning | Doyle, Mensah, Chen | 3 | | OP20 Milling | Doyle, Garza, Esposito | 3 | | OP30 Drilling | Garza, Alvarez, Fofana | 3 | | OP40 Deburr | Haddad, Jensen, Ito | 3 | | OP50 Wash | Haddad, Jensen, Kowalski | 3 | | OP60 CMM inspection | Mensah, Alvarez, Brooks | 3 | | Line 4 night total | | 18 of 18 qualified slots = 100%. 3x3 rule met. | OP60 last assessed: Mensah 2026-02-11 · Alvarez 2025-03-04 · Brooks 2026-05-20 Requalification required every 12 months (procedure QP-7.2). Toledo plant coverage (all lines, all shifts): 94% [3] Teams message from Dwayne Pruitt, Line 4 night supervisor, Toledo honestly we're getting by but I'm moving the same few people around every night. Doyle went to days Sept 15 and Mensah quit Oct 3. if anyone calls in on CMM we stop shipping to the customer, Brooks is basically doing every check. can we get more people CMM trained? and I need OT approved for Dec [4] Minutes: Joint Union–Management Training Committee, Toledo (Oct 28) Present: 3 union representatives, 3 management representatives (including L. Ortega). 1. Article 14.3 of the collective agreement commits the company to 40 hours of paid skills training per hourly employee per year. Toledo has 950 hourly employees (commitment: 38,000 hours for 2026). 2. Management reported 31,200 training hours delivered year to date: 82% of the annual commitment. Management described the plant as on track. 3. The union noted that the 31,200 hours include 7,600 hours of new-hire orientation. The union's position is that Article 14.3 covers skills training and excludes orientation. Management did not dispute this. 4. The union raised concern that night-shift employees cannot get to training held on day shift. HR reported 210 employees with fewer than 10 hours this year. 5. Action: management to provide hours per employee, by shift, by December 1. [5] Email from Carla Whitfield, Training Manager, Florence Subject: URGENT retraining Line 2 torque Luis, we had a customer complaint this week. They found 40 brackets with the fastener torqued to 25 Nm. The plant manager wants every Line 2 operator retrained on work instruction WI-L2-07 by end of next week. Can you approve the hours? I've attached the WI. (Attached: WI-L2-07, revision C, dated August 4, 2026: "Torque bracket fastener to 25 Nm ± 2.") (Forwarded below by quality, Oct 21: "ECN-2291 released Oct 20: bracket fastener torque changed from 25 Nm to 32 Nm for the new model year. Work instructions to be updated by process engineering.") [6] Plant summary: Silao From: Training coordinator, Silao All 9 lines meet the 3x3 rule. Plant coverage 97%. Note: the matrix covers shifts 1 and 2. Shift 3 runs lines 1 to 4 with a reduced crew. [7] Plant summary: Apodaca Apodaca · Skills coverage, November Coverage (Level 2+): 99% Key: Level 1 = in training · Level 2 = trained, works under supervision · Level 3 = qualified to work unsupervised · Level 4 = can train othersI lead technical training at an automotive components manufacturer with about 7,200 people in four plants. Below are this month's skills-matrix exports, plant summaries, a supervisor's message, an email from one of our plants and the minutes of our union training committee. I need a one-page pre-read for our plant leadership council, which meets on November 19.
Please read everything carefully and give me:
Be skeptical. Where a summary, total or percentage says more than the details underneath support, go with the details and say so. Check that numbers measure the same thing before you compare or add them, and check dates: people who have moved or left, and qualifications past their renewal date, don't count. If something is missing, say it's missing. Tell me which input each point comes from. Keep it to one page.
Subject: Council pre-read for Nov 19
Luis, same as last month please. One page, coverage % by plant, and anything the council needs to decide. Last month we showed Toledo 94, Silao 97, Florence 91, Apodaca 99, so if those moved, tell us why.
Thanks, Ines
Narrowgate Skills Matrix · Toledo · Line 4 (machining) · Shift: Night Matrix last updated: 2026-06-30 · Export run: 2026-11-10
| Station | Qualified operators (Level 3+) | Count |
|---|---|---|
| OP10 Turning | Doyle, Mensah, Chen | 3 |
| OP20 Milling | Doyle, Garza, Esposito | 3 |
| OP30 Drilling | Garza, Alvarez, Fofana | 3 |
| OP40 Deburr | Haddad, Jensen, Ito | 3 |
| OP50 Wash | Haddad, Jensen, Kowalski | 3 |
| OP60 CMM inspection | Mensah, Alvarez, Brooks | 3 |
| Line 4 night total | 18 of 18 qualified slots = 100%. 3x3 rule met. |
OP60 last assessed: Mensah 2026-02-11 · Alvarez 2025-03-04 · Brooks 2026-05-20 Requalification required every 12 months (procedure QP-7.2).
Toledo plant coverage (all lines, all shifts): 94%
honestly we're getting by but I'm moving the same few people around every night. Doyle went to days Sept 15 and Mensah quit Oct 3. if anyone calls in on CMM we stop shipping to the customer, Brooks is basically doing every check. can we get more people CMM trained? and I need OT approved for Dec
Present: 3 union representatives, 3 management representatives (including L. Ortega).
Subject: URGENT retraining Line 2 torque
Luis, we had a customer complaint this week. They found 40 brackets with the fastener torqued to 25 Nm. The plant manager wants every Line 2 operator retrained on work instruction WI-L2-07 by end of next week. Can you approve the hours? I've attached the WI.
(Attached: WI-L2-07, revision C, dated August 4, 2026: "Torque bracket fastener to 25 Nm ± 2.")
(Forwarded below by quality, Oct 21: "ECN-2291 released Oct 20: bracket fastener torque changed from 25 Nm to 32 Nm for the new model year. Work instructions to be updated by process engineering.")
From: Training coordinator, Silao All 9 lines meet the 3x3 rule. Plant coverage 97%. Note: the matrix covers shifts 1 and 2. Shift 3 runs lines 1 to 4 with a reduced crew.
Apodaca · Skills coverage, November Coverage (Level 2+): 99% Key: Level 1 = in training · Level 2 = trained, works under supervision · Level 3 = qualified to work unsupervised · Level 4 = can train others