genimage2

name	genImage2
description	Generate images through the current Codex-compatible Responses gateway using the image_generation tool. Use when the user wants to create an image with the current configured gateway, save a PNG in the current working directory, or override the current model for a one-off image generation request.

genImage2

Use this skill when the user wants to generate an image through the current Codex gateway configuration instead of calling the official OpenAI endpoint directly.

Do not hardcode credentials, gateway URLs, or machine-specific paths. Read them at runtime from the current Codex configuration:

• the active Codex config file for base_url and the current default model
• the active Codex auth file for OPENAI_API_KEY

Use the bundled script:

scripts/generate_response_image.sh \
  --prompt "生成一张仙侠女主图片"

Override the model when the user asks for a specific one:

scripts/generate_response_image.sh \
  --prompt "Generate a cinematic poster of a gray tabby cat hugging an otter" \
  --model gpt-5.5

Write to a specific file when the user wants a named output:

scripts/generate_response_image.sh \
  --prompt "生成一张国风海报" \
  --output ./poster.png

Notes:

• This skill uses POST {base_url}/responses with tools: [{"type":"image_generation"}].
• If the request returns a base64 image result, the script decodes it into a local PNG.
• Without --output, the script writes the PNG into the current working directory, which is typically the current conversation directory.
• If the current gateway responds slowly, allow a longer timeout instead of assuming the request failed immediately.