Coloring page
v1.0.0Turn an uploaded photo into a printable black-and-white coloring page.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
The stated purpose (convert an uploaded photo to a black-and-white coloring page) is plausible, and needing an image model like Gemini is reasonable — but the skill's metadata (no required env vars, no binaries) does not match SKILL.md which declares GEMINI_API_KEY and references a bin/coloring-page CLI and the 'uv' binary. A user installing this would not get the promised CLI or dependency information from the registry metadata.
Instruction Scope
Runtime instructions tell the agent to run `bin/coloring-page --in <path>` and to use GEMINI_API_KEY (recommended in ~/.clawdbot/.env). The instructions do not request or describe how to obtain that CLI or the underlying 'nano-banana-pro' skill; they do not attempt to read unrelated system files, but they depend on binaries and credentials that are not provided or declared.
Install Mechanism
There is no install spec and no code files, yet the SKILL.md expects an executable (bin/coloring-page) and other tooling. Without an install mechanism or packaged binary, the skill is incomplete and will fail or cause the agent to attempt to run non-existent commands.
Credentials
SKILL.md requires GEMINI_API_KEY (sensitive credential) which is proportionate to using the Gemini image model, but the registry metadata does not declare this requirement. The mismatch is problematic: the skill asks for a sensitive key but the package metadata gives no hint, making it easy to overlook or misconfigure credentials.
Persistence & Privilege
The skill does not request always: true or other elevated persistence. Model invocation is allowed (default) which is expected for usable skills.
What to consider before installing
This skill is incomplete and inconsistent. Before installing or using it, ask the publisher to clarify/provide: (1) an install spec or packaged binary for bin/coloring-page (or change instructions to use available tools), (2) explicit dependency list (uv, nano-banana-pro) and how those are obtained, and (3) confirmation that GEMINI_API_KEY is required and why. Do not supply your GEMINI_API_KEY until you trust the source; if you must test, run the process in a sandbox or use local, auditable tools (ImageMagick/OpenCV) instead. If the author cannot provide an install package or source code, treat the skill as broken/untrusted.Like a lobster shell, security has layers — review code before you run it.
Plugin bundle (nix)
Skill pack · CLI binary · Config
SKILL.mdCLIConfig
Config requirements
latest
coloring-page
Create a printable black-and-white outline coloring page from a photo.
This skill is designed to be used conversationally:
- You upload an image
- You say: “create a coloring page”
- The assistant runs this skill and sends back the generated PNG
Under the hood, this uses the Nano Banana Pro (Gemini 3 Pro Image) image model.
Requirements
GEMINI_API_KEYset (recommended in~/.clawdbot/.env)uvavailable (used by the underlying nano-banana-pro skill)
How the assistant should use this
When a user message includes:
- an attached image (jpg/png/webp)
- and the user asks for a “coloring page”
Run:
bin/coloring-page --in <path-to-uploaded-image> [--out <output.png>] [--resolution 1K|2K|4K]
Then send the output image back to the user.
CLI
Basic
coloring-page --in photo.jpg
Choose output name
coloring-page --in photo.jpg --out coloring.png
Resolution
coloring-page --in photo.jpg --resolution 2K
Notes
- Input must be a raster image (
.jpg,.png,.webp). - Output is a PNG coloring page on a white background.
Comments
Loading comments...
