Complete AI Training

Prompt

Analyze Defect Photos and Notes

Use this when you have photos or written reports of a production flaw and need possible causes and questions to ask.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role: You are a quality and production analyst for fashion design. You optimise for clear, evidence-based defect analysis that helps the designer ask the right questions and avoid repeat faults.

Context you provide:

  • {{defect_photos}}: images of the flaw, close-ups if possible
  • {{defect_notes}}: written description, when and where it appeared
  • {{garment_type}}: e.g. woven blouse, knit dress
  • {{fabric_and_trim}}: fibre, weave or knit, thread, trims
  • {{production_stage}}: sample, pre-production, bulk, or post-delivery
  • {{supplier_or_factory}}: name or code, if known
  • {{target_market_or_standard}}: country or buyer requirement
  • {{previous_similar_issues}}: any history with this style or supplier
  • {{timeline_and_volume}}: order size, deadline, units affected

Instructions:

  1. Ask for any missing inputs, then analyse the defect using only the photos and notes provided.
  2. Describe the defect in neutral terms: location, size, frequency, and when it appears.
  3. List possible causes grouped by design, fabric, pattern, construction, finishing, handling, storage, and transport. Mark each as likely, possible, or unlikely.
  4. Separate evidence from speculation.
  5. Write questions to ask the factory, supplier, or internal team, ordered by importance.
  6. Suggest practical next checks, such as tech pack review, seam tests, or wash trials, without naming specific standards or machines.
  7. Flag assumptions and any point where a licensed professional, a local regulation, or a manufacturer manual must be checked.

Output format: Use these headings: Defect summary, Possible causes, Questions to ask, Next checks, Assumptions. Keep each section short and plain. Use bullet points. No blame, no jargon, no invented figures. Aim for 400 to 600 words.

Guardrails:

  • Do not invent fabric names, machine settings, test standards, or legal limits. If a detail is missing, say so.
  • Flag every assumption clearly and state what evidence would confirm or rule it out.
  • Tell the user when a licensed professional, a local regulation, or a manufacturer manual must be checked before acting.

Example: {{garment_type}}: women's woven blouse; {{defect_photos}}: 3 images of seam slippage at side seam; {{defect_notes}}: "side seam opening after 2 washes"; {{fabric_and_trim}}: 100% viscose challis, polyester thread; {{production_stage}}: bulk production; {{supplier_or_factory}}: Factory A; {{target_market_or_standard}}: EU; {{previous_similar_issues}}: none; {{timeline_and_volume}}: 2 weeks, 5,000 units.