← Back to Skills Marketplace
NBP
by
Youthzenith
· GitHub ↗
· v1.0.0
· MIT-0
84
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install nanobananapro
Description
Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-image; 1K...
Usage Guidance
This skill's code and README clearly implement a Gemini (Nano Banana Pro) image generator and require a GEMINI_API_KEY and the 'uv' runner plus Python packages (google-genai, pillow). The registry metadata, however, does not declare the GEMINI_API_KEY or required binaries — that's the main inconsistency. Before installing or running: 1) Verify the publisher/owner (ownerId mismatch between inputs and _meta.json). 2) Expect to provide a Google/Gemini API key (GEMINI_API_KEY) — only give this to code you trust; do not paste it into public chat. 3) Ensure you have 'uv' and the Python dependencies installed in a safe environment. 4) If you rely on the skill manifest for security review, ask the publisher to update the registry metadata to explicitly list GEMINI_API_KEY and required binaries/dependencies. If you cannot verify the publisher or do not want to expose your API key, do not install/use the skill.
Capability Analysis
Type: OpenClaw Skill
Name: nanobananapro
Version: 1.0.0
The skill is a legitimate implementation for generating and editing images using Google's Gemini API. The Python script `scripts/generate_image.py` uses the official `google-genai` and `pillow` libraries to handle image processing and API communication, with no evidence of data exfiltration, unauthorized file access, or malicious execution. The instructions in `SKILL.md` are well-structured for an AI agent and do not contain any prompt-injection attacks or deceptive directives.
Capability Tags
Capability Assessment
Purpose & Capability
The skill's purpose (Nano Banana Pro / Gemini image generation) matches the code and SKILL.md: the script calls google.genai and uses image generation. However the registry metadata declares no required environment variables or binaries while both the SKILL.md and the script require a GEMINI_API_KEY and expect the 'uv' runtime. Also the script lists python dependencies (google-genai, pillow) in its header comments but the skill manifest provides no install spec. There is also a mismatch between ownerId in the provided registry metadata and _meta.json, which is suspicious/inconsistent.
Instruction Scope
SKILL.md instructs the agent to run the included Python script to generate or edit images and to supply an API key either via --api-key or GEMINI_API_KEY. The instructions do not ask the agent to read unrelated files, exfiltrate data, or contact unexpected endpoints; they limit I/O to image files and the Gemini API. The note to run from the user's working directory and to not read images back is consistent with the stated purpose.
Install Mechanism
There is no install spec (instruction-only skill), which is low risk. The included script, however, depends on external Python packages (google-genai, pillow) and the 'uv' runner; these are only listed in comments and in SKILL.md preflight but are not enforced or declared in the registry. That discrepancy is noteworthy because required packages must be present for the skill to work.
Credentials
The script and SKILL.md explicitly require a GEMINI_API_KEY (or passing --api-key). Yet the skill's declared required environment variables and primary credential fields are empty. Requesting an API key for the vendor service is reasonable for this functionality, but failing to declare it in the manifest is an incoherence that reduces transparency and makes it harder to audit what secrets the skill will use.
Persistence & Privilege
The skill does not request persistent 'always' inclusion and does not modify other skills or global agent settings. It runs as an on-demand script and only requires the Gemini API key for operation.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install nanobananapro - After installation, invoke the skill by name or use
/nanobananapro - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of Nano Banana Pro skill for image generation and editing via Gemini 3 Pro Image API.
- Supports both text-to-image and image-to-image workflows, with resolutions up to 4K.
- Adds workflow guidance for iterative image creation (draft, iterate, final).
- Automates filename generation with timestamp and descriptive naming.
- Handles API key sourcing from argument or environment variable.
- Includes detailed usage instructions, prompt handling tips, and error troubleshooting.
Metadata
Frequently Asked Questions
What is NBP?
Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-image; 1K... It is an AI Agent Skill for Claude Code / OpenClaw, with 84 downloads so far.
How do I install NBP?
Run "/install nanobananapro" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is NBP free?
Yes, NBP is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does NBP support?
NBP is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created NBP?
It is built and maintained by Youthzenith (@youthzenith); the current version is v1.0.0.
More Skills