Nano Image is an agent skill that adds text-to-image generation capabilities by connecting to Google’s Gemini 2.0 Flash image generation pipeline. When an agent invokes the skill with a text prompt, the skill handles API communication, base64 decoding of the response, and saving the resulting file to a local directory at ~/.claude-image-gen. Output files are standard PNG or JPG format, making them immediately usable in web projects, presentations, or other workflows.
The skill uses Python’s urllib standard library for HTTP communication, which keeps the dependency footprint small. It creates the output directory automatically if it does not exist, so no manual setup of the file system is required before first use. Built-in rate-limit handling is included to stay within Google’s free-tier quotas and avoid API access issues.
According to the description, the skill targets diverse visual styles including photorealism, vector art, 3D renders, and logos. This makes it applicable to tasks such as generating UI mockup placeholders, concept art for presentations, or illustrative assets for written content. No package registry entry, version number, or license is published for this skill at this time, so integrators should verify compatibility with their environment before deploying.