• User asks to generate an illustration, diagram, or concept image
• User is writing an article and needs visual explanations for concepts or workflows
• User explicitly asks for AI-generated images

1. What to illustrate — the concept, architecture, flow, or scene
2. Language — default to English for both prompt and text in image. Only use other languages if the user explicitly requests it
3. Save location — see "Output Path" section below
4. Style/color preferences — if user has specific needs, otherwise use defaults

Use a clean, modern color palette with soft tones. Minimalist flat illustration style with clear visual hierarchy. Professional and polished look suitable for technical blog articles. No photorealistic rendering. No excessive gradients or shadows.

• Be specific and descriptive about the visual elements
• For technical concepts: describe the components, their relationships, and layout
• For architecture diagrams: list the layers/components and how they connect
• For flow diagrams: describe the steps and direction of flow
• If text labels are needed in the image, spell them out explicitly in the prompt
• Default language is English. Only write the prompt and request in-image text in another language if the user specifically asks

1. Check where existing images in that article are stored (look for

   ![

   image references in the

   .md

   file)
2. Save the new image in the same directory as the existing images
3. Use a descriptive filename that matches the naming convention of existing images

![](images/architecture-overview.png)

images/

1. Look for existing image directories:

   images/

   ,

   assets/

   ,

   static/

   ,

   img/

   ,

   figures/

2. Save in the most appropriate existing directory
3. If none exists, create an

   images/

   directory at the project root or under the relevant content directory

1. Save to the current working directory
2. Use a descriptive filename:

   concept-name-illustration.png

• Use lowercase with hyphens:

  vector-search-architecture.png

• Be descriptive: prefer

  rag-pipeline-overview.png

  over

  image1.png

• Match existing naming patterns in the project if any

1. Read the image file to visually verify it matches the user's request
2. If the result is not satisfactory, refine the prompt and regenerate
3. If the image will be inserted into a markdown file, suggest the markdown syntax:

   ![alt text](relative/path/to/image.png)