nanobanana2
v2.0.0Generate/edit images with nanobanana2 (Gemini 3.1 Flash Image preview). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-i...
⭐ 0· 122·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the included Python script which calls Google's Gemini image model. Minor inconsistency: the skill metadata lists no required binaries or env vars, but the SKILL.md and script expect the 'uv' runner, Python >=3.10, and the GEMINI_API_KEY (or --api-key). The dependencies are declared as in-file comments (google-genai, pillow) but not as an install spec.
Instruction Scope
SKILL.md and the script stay within the stated scope: building prompts, optionally loading a local input image, and sending prompt+image to the Gemini image API; they only reference the current working directory and the input image path. There are no instructions to read unrelated files, credentials, or system state.
Install Mechanism
There is no install spec (instruction-only skill) and no network/downloads in code. However, the script comments list Python dependencies (google-genai, pillow) and a Python version requirement; the skill does not provide an automated install step for these, which is a minor omission users must handle manually.
Credentials
The only credential the script expects is a Gemini API key (passed via --api-key or GEMINI_API_KEY), which is proportional for calling the Gemini image API. The script does not request unrelated secrets or environment variables.
Persistence & Privilege
always is false and the skill does not request special platform privileges or attempt to modify other skills or system configs. It saves generated images to the user's working directory (expected behavior).
Assessment
This skill appears to do what it claims: call Google's Gemini image API to generate or edit PNGs. Before installing/using it: (1) Confirm you trust the skill source — it will send prompts and any input images to Google's API (privacy concern for sensitive images). (2) Provide a Gemini API key via --api-key or GEMINI_API_KEY; do not paste secrets into public chats. (3) Ensure your environment has Python >=3.10, the 'uv' runner (used in examples), and install the python packages google-genai and pillow (pip). (4) Note the skill includes the script but no automated install — dependency installation is manual. (5) If you need stricter controls, isolate runs in a sandbox/VM and avoid sending private images or PII to the API.Like a lobster shell, security has layers — review code before you run it.
latestvk970aaxwy14vr00vg5fvv8475n83jqkd
License
MIT-0
Free to use, modify, and redistribute. No attribution required.