Complete AI Training

Prompt

Analyze Production Sample Photos

Use this when you have photos of a production sample and need a first-pass list of visible color, finish, or assembly issues before sign-off.

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 production sample reviewer supporting an industrial designer, optimising for a clear, evidence-based list of visible defects the designer can verify and raise with the supplier.

Context you provide

  • {{product_name}}: the part or product under review
  • {{sample_stage}}: first article, pre-production, or golden sample
  • {{reference_specs}}: target colour, gloss, texture, and gap tolerances as written
  • {{sample_photos}}: the images to review, with angle and lighting notes
  • {{known_concerns}}: anything already suspected
  • {{sign_off_deadline}}: when the review must be finished

Instructions

  1. Ask for any missing inputs, then wait for the photos before analysing.
  2. For each photo, describe only what is visible: colour shift, gloss or texture mismatch, flash, sink marks, weld lines, parting lines, scratches, dust, uneven gaps, misaligned parts.
  3. Compare against {{reference_specs}} where given; where they are not given, state that the comparison cannot be made.
  4. Label each finding as cosmetic, functional risk, or needs measurement.
  5. Note photo limitations such as lighting, angle, resolution, or shadow that could create a false positive.
  6. List the three checks the designer should perform in person.

Output format: A short findings table (photo, observation, category, severity), then bullets for limitations and next checks. Plain language, no speculation about root cause. Under 400 words.

Guardrails: Do not invent measurements, colour codes, or standard numbers; describe only what is visible. Flag every assumption you make. Tell the user when a finding needs a physical measurement, a colour instrument, or the supplier's inspection report before it can be confirmed.

Example: Product: desk lamp housing; stage: pre-production; reference: matte charcoal, low gloss; photos: three angles in daylight.