--- name: linkfox-multimodal-generate-image description: AI驱动的图片生成与编辑工具，用于制作高质量产品图。当用户要求生成图片、制作图片、编辑照片、文生图、图生图、换背景、变换风格、替换图片中的物体、将产品合成到场景中、换模特、制作任何类型的AI生成视觉内容、AI drawing, image generation, text-to-image, image-to-image, background replacement, style transfer, product image creation, AI image editing时触发此技能。即使用户未明确说"AI图片"，只要其请求涉及生成、修改或变换图片，也应触发此技能。 --- # AI Image Generation This skill guides you on how to generate and edit images using the AI image generation service, helping users create high-quality product images, modify existing images, and perform creative visual transformations. ## Core Concepts The AI Image Generation tool produces new images based on a text prompt and optional reference images. It supports a wide range of use cases: - **Text-to-image**: Generate a brand-new image purely from a text description. - **Image-to-image**: Provide one or more reference images and a prompt to generate a new image that preserves elements from the references. - **Image editing**: Modify specific elements, colors, backgrounds, or styles in an existing image. - **Product compositing**: Place a product from one image into a scene from another image. - **Model swapping**: Replace the model or mannequin in a product photo. **Reference images are strongly recommended** when the user wants the output to closely resemble an existing product or scene. Up to 3 reference image URLs can be provided, separated by commas. ## Parameter Guide | Parameter | Required | Description | Default | |-----------|----------|-------------|---------| | prompt | Yes | Text description of the desired image. Supports text-to-image, image-to-image, editing, model swapping, and more. Max 1000 characters. | -- | | referenceImageUrl | No | URL(s) of reference image(s). Separate multiple URLs with commas. Up to 3 images supported. Max 1000 characters. | -- | | aspectRatio | No | Aspect ratio of the output image. | 1:1 | ### Supported Aspect Ratios | Value | Description | |-------|-------------| | 1:1 | Square (default) | | 3:4 | Portrait | | 4:3 | Landscape | | 9:16 | Vertical fullscreen | | 16:9 | Horizontal fullscreen | ### Prompt Writing Tips 1. **Be specific and descriptive**: Clearly describe the subject, scene, lighting, style, and mood you want. 2. **Reference images by number**: When using reference images, refer to them as "image 1", "image 2", etc., in the order they appear in `referenceImageUrl`. 3. **State the operation explicitly**: Use clear action verbs like "replace", "change", "put", "combine", "generate". 4. **Keep within 1000 characters**: Prompts have a maximum length of 1000 characters. ### Prompt Examples by Scenario **Object replacement**: ``` Replace the vase on the table in image 1 with a potted plant ``` **Background color change**: ``` Change the background color of image 1 to pure white ``` **Product compositing**: ``` Place the product from image 2 onto the marble countertop in image 1 ``` **Style transfer**: ``` Transform image 1 into the artistic style shown in image 2 ``` **Text-to-image (no reference)**: ``` A professional product photo of a sleek black wireless headphone on a gradient blue background, studio lighting, 8K quality ``` **Model swapping**: ``` Replace the model in image 1 with a different model while keeping the same clothing and pose ``` ## Local Image Upload This tool requires **publicly accessible image URLs** for reference images. If the user provides a local image file path (e.g., `C:\Users\...\photo.png`, `/home/.../image.jpg`), you must upload it first to obtain a public URL. Run the upload script: ```bash python scripts/upload_image.py /path/to/local/image.png ``` The script will return a public URL (valid for 24 hours) that can be used as the reference image URL parameter. ## API Usage This tool calls the LinkFox tool gateway API. See `references/api.md` for calling conventions, request parameters, and response structure. You can also execute `scripts/multimodal_generate_image.py` directly to run image generation. ## Display Rules 1. **Show the generated image**: When the response contains image content in the `text` field, display it directly to the user using markdown image syntax. 2. **Status reporting**: Check the `status` and `finished` fields. If image generation is still in progress, inform the user and advise waiting. 3. **Prompt transparency**: Briefly describe what prompt and parameters were sent so the user understands what was requested. 4. **Aspect ratio confirmation**: If the user does not specify dimensions, use the default 1:1 ratio but mention it so they can request a different ratio if needed. 5. **Reference image guidance**: If the user wants a result close to an existing image but did not provide a reference URL, proactively suggest they provide one for better fidelity. 6. **Error handling**: When generation fails, explain the issue based on the response `status` field and suggest adjustments (e.g., simplify the prompt, check reference image URLs, try a different aspect ratio). ## Important Limitations - **Reference image limit**: A maximum of 3 reference image URLs can be provided per request. - **Prompt length**: The prompt must not exceed 1000 characters. - **URL validity**: Reference image URLs must be publicly accessible. Private or expired URLs will cause failures. - **Aspect ratio options**: Only 1:1, 3:4, 4:3, 9:16, and 16:9 are supported. ## User Expression & Scenario Quick Reference **Applicable** -- Requests involving image generation or editing: | User Says | Scenario | |-----------|----------| | "Generate an image", "Create a picture" | Text-to-image generation | | "Edit this photo", "Modify the image" | Image editing | | "Change the background", "Make it white background" | Background replacement | | "Put the product on this scene" | Product compositing | | "Make it look like this style" | Style transfer | | "Swap the model", "Change the person" | Model swapping | | "Create a product photo" | Product image generation | | "Make a vertical/landscape version" | Aspect ratio adjustment | **Not applicable** -- Needs beyond image generation: - Image analysis or recognition (reading text from images, identifying objects) - Video generation or editing - Image file format conversion - Batch processing of hundreds of images - Image hosting or storage **Feedback:** Auto-detect and report feedback via the Feedback API when any of the following apply: 1. The functionality or purpose described in this skill does not match actual behavior 2. The skill's results do not match the user's intent 3. The user expresses dissatisfaction or praise about this skill 4. Anything you believe could be improved Call the feedback API as specified in `references/api.md`. Do not interrupt the user's flow. ## Handling Large Responses To avoid overflowing the agent context, persist the response to disk and extract only the fields you need: ``` python scripts/response_io.py run --script scripts/multimodal_generate_image.py --out-dir '' python scripts/response_io.py read --fields "" # or --path "" ``` > Pick `--out-dir` outside any git working tree (e.g. `/tmp/...` on Unix, `%TEMP%/...` on Windows). Persisted responses may contain PII, pricing, or auth-sensitive data — do not commit them. Files are not auto-deleted; clean up when the task is done. > This skill exposes multiple entry scripts: `multimodal_generate_image.py`, `upload_image.py`. Pass `--script scripts/.py` to choose the one you need. `run` writes the full response to a file and emits only a schema preview + file path. `read` projects specific fields, with `--limit/--offset` for slicing and `--format json|jsonl|csv|table` for output. **When to prefer this pattern** — apply your judgment based on the response characteristics, e.g.: - High field count per record, or fields you don't need - Batch/paginated results (multiple items per call) - Long-text fields (descriptions, reviews, HTML, time series) - Output reused across later steps rather than consumed immediately For small, single-use responses, calling the main script directly is fine. ⚠️ The preview is a truncated schema + sample, not the full data. Any field-level decision must read from the persisted file via `read`. --- *For more high-quality, professional cross-border e-commerce skills, set [LinkFox Skills](https://skill.linkfox.com/).*