Clay Robot Watering a Plant | Follow-Along Exercise
Practice creating a clay robot watering a plant: start from concrete inputs, then complete, check and save your own version step by step.
できあがるもの
A saveable visual plan or image, a prompt and a revision log.
必要なもの
- A chat, image or video tool suited to the task (PromptKite does not provide model credits)
- Your own input materials and a folder for saving results
手順
- Define the inputs for this exercise
The demo inputs for this exercise are: Subject: a round, milky-white robot; Palette: light blue and warm orange. First confirm whether these conditions fit your situation; replace “paste your material” in the example with real material. Write the goal as a deliverable you can check against.
- Run it once for your scenario
Paste the filled-in example prompt below into a tool that supports image generation. For real products, upload your own reference photos and don't pass off concept images as real photos; for layout-planning tasks, get the layout plan first. Focus of this task: the robot has two black dot eyes and round orange ears, and is watering a green seedling in a terracotta pot.
- Check against the criteria and revise
Check the output against “Water direction, how the hand grips the watering can, and the pot's position; zoom in to check edges, then verify the aspect ratio and space for layout.” and note the most obvious problem. Change only one of subject scale, lighting or white space first, then regenerate and compare, instead of swapping all the style words at once.
- Save and review
Save the final result, the prompt, the source of your input material and one reason for a change. Name files with the date plus the task name; save reusable versions to your PromptKite Prompt Pocket. Next time, change one condition and try again, noting whether the result still holds.
よくある失敗
- The input is still placeholder text, so the output looks empty — Replace placeholders such as “paste your material” in the example with real material; state clearly that missing facts are unknown instead of asking the model to fill them in.
- The result doesn't meet the standard for this scenario — Check the output against “Water direction, how the hand grips the watering can, and the pot's position; zoom in to check edges, then verify the aspect ratio and space for layout.” and note the most obvious problem. Change only one of subject scale, lighting or white space first, then regenerate and compare, instead of swapping all the style words at once.
練習用テンプレート
Create an original visual of {subject}, with a main palette of light blue and warm orange. The robot has two black dot eyes and round orange ears, and is watering a green seedling in a terracotta pot. Handmade clay texture, rounded but with visible joints, soft studio lighting, the pot and robot touching the ground, and a large area of clean white space above. Use a 4:5 portrait format with a clear hierarchy, details serving the subject, and space reserved for later typography. Don't generate text, logos, signatures or watermarks; this is a concept example and does not represent a real product photo. If I provide a real product reference, follow its structure and colors; if information is missing, don't invent product selling points. After generating, check the image edges, object connections, perspective, and light and shadow; when adjusting, change only one factor at a time.This page is an original follow-along exercise; the full workflow has not yet been tested end to end. The image reuses an actual GPT-generated example of this prompt from PromptKite and does not represent a finished poster layout or other deliverable.