Skills nano-image-generator

📦

nano-image-generator

Name: nano-image-generator
Author: lxfater

Content revision r1 High Risk 🌐 Network access📁 Filesystem access🔑 Env variables⚙️ External commands

Create Gemini Image Assets

Teams need quick visual assets without leaving their coding workflow. This skill uses a local Python script to generate images with Gemini and save them into project folders.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Stop for confirmation before installing.

Review the plan and obtain explicit user consent before changing files.

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "nano-image-generator" from https://skillstore.io/skills/lxfater-nano-image-generator.md and its manifest at https://skillstore.io/api/skills/lxfater-nano-image-generator/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/lxfater-nano-image-generator.md Signed manifest GET /api/skills/lxfater-nano-image-generator/manifest Signed lockfile GET /api/skills/lxfater-nano-image-generator/lockfile

Test it

Using "nano-image-generator". Create a square app icon for a weather app with a simple cloud and sun symbol.

Expected outcome:

A saved square image file suitable for an app icon draft, using the selected output folder.

Using "nano-image-generator". Create a wide launch banner for a developer tool in a clean editorial style.

Expected outcome:

A landscape banner image with a generated file path returned after the script completes.

Using "nano-image-generator". Use this character reference and place the same character in a warm cafe scene.

Expected outcome:

A new image that attempts to preserve the referenced character appearance in a different setting.

Security Audit

High Risk

v7 • 7/5/2026 Open versioned report

The skill provides a legitimate Gemini-backed image generation workflow, but it has meaningful security concerns. The main confirmed issues are hardcoded API key setup, external transfer of prompts and reference images, local Python command execution, and unrestricted output-path writes. Documentation-only URLs, Markdown formatting, magic-byte checks, and response image base64 decoding were adjudicated as false positives.

Files scanned

685

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (3)

High

README.md:27-47 README_CN.md:27-47 scripts/generate_image.py:37-45

Hardcoded API Key Setup Workflow

The README files and Python script instruct users to replace a source-code placeholder with a real Gemini API key. This creates a high chance of committing or sharing a live credential.

The documentation explicitly says to replace the placeholder with an actual API key, and the script returns that literal value. This is direct evidence of unsafe credential storage by design.

Medium

scripts/generate_image.py:119-125 scripts/generate_image.py:143-147

Reference Images Sent to External API

The script reads each reference image, base64-encodes it, includes it in the request payload, and posts it to the Gemini API. Users may expose private images if they provide sensitive local files.

The code appends reference image bytes as inlineData and then sends the payload through urllib.request.urlopen. This is expected functionality, but it is a clear external data transfer.

Medium

scripts/generate_image.py:254-268

Unrestricted Output Path Write

The output path is provided by the caller, parent directories are created automatically, and image bytes are written to that path. A mistaken or manipulated path could overwrite files the process can access.

The script directly converts args.output into a Path, creates parent directories, and writes bytes. The forced image extension limits some abuse, but path scope is otherwise unrestricted.

Capability review items (30)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

High

README_CN.md:29

Generic API/secret keys

编辑 `scripts/generate_image.py`,找到 `get_api_key()` 函数(约第 37 行):

The setup instructions direct users to edit the script function that returns the API key. This encourages storing a live credential in source code.

High

README_CN.md:32

Generic API/secret keys

def get_api_key():

The documented get_api_key function is the credential storage location. The surrounding setup text tells users to replace a placeholder with a real key.

High

README_CN.md:40

Generic API/secret keys

return "YOUR_GEMINI_API_KEY_HERE" # <-- Replace this with your API key

The snippet returns an API key placeholder from source code. Replacing it with a real key would hardcode a secret in the repository.

High

README_CN.md:43

Generic API/secret keys

将 `YOUR_GEMINI_API_KEY_HERE` 替换为你的实际 API key:

The documentation explicitly tells users to replace the placeholder with an actual API key. This is insecure credential handling.

High

README.md:29

Generic API/secret keys

Edit `scripts/generate_image.py`, find the `get_api_key()` function (around line 37):

The setup instructions direct users to edit the script function that returns the API key. This encourages storing a live credential in source code.

High

README.md:32

Generic API/secret keys

def get_api_key():

The documented get_api_key function is the credential storage location. The surrounding setup text tells users to replace a placeholder with a real key.

High

README.md:40

Generic API/secret keys

return "YOUR_GEMINI_API_KEY_HERE" # <-- Replace this with your API key

The snippet returns an API key placeholder from source code. Replacing it with a real key would hardcode a secret in the repository.

High

README.md:43

Generic API/secret keys

Replace `YOUR_GEMINI_API_KEY_HERE` with your actual API key:

The documentation explicitly tells users to replace the placeholder with an actual API key. This is insecure credential handling.

High

scripts/generate_image.py:5

Generic API/secret keys

Requires: Set your GEMINI_API_KEY in the get_api_key() function below.

The file-level instructions require placing the Gemini API key into the function below. This promotes hardcoded credential storage.

High

scripts/generate_image.py:37

Generic API/secret keys

def get_api_key():

The get_api_key function is used as the credential source for requests. Its implementation returns a literal string from the source file.

High

scripts/generate_image.py:45

Generic API/secret keys

return "YOUR_GEMINI_API_KEY_HERE" # <-- Replace this with your API key

The function returns a placeholder meant to be replaced by a real API key. Hardcoding a real secret here risks accidental disclosure.

High

scripts/generate_image.py:101

Generic API/secret keys

api_key = get_api_key()

The script retrieves the API key from the hardcoded function before making the request. This ties runtime authentication to source-stored credentials.

High

scripts/generate_image.py:103

Generic API/secret keys

if api_key == "YOUR_GEMINI_API_KEY_HERE":

The placeholder comparison confirms the expected setup is replacing a source-code literal with a real key. That setup pattern is insecure.

High

scripts/generate_image.py:105

Generic API/secret keys

print("Edit the get_api_key() function and replace YOUR_GEMINI_API_KEY_HERE", file=sys.stderr)

The error message instructs users to replace the placeholder in source code. This reinforces unsafe credential handling.

High

scripts/generate_image.py:108

Generic API/secret keys

url = f"{API_BASE}/{MODEL_ID}:generateContent?key={api_key}"

The API key is placed in the request URL query string. Query-string credentials can leak through logs, traces, or error reports.

Medium

scripts/generate_image.py:144

Python HTTP libraries

req = urllib.request.Request(url, data=data, headers=headers, method="POST")

The script creates a POST request containing the prompt and any encoded reference images. This sends user-provided content to an external API.

Medium

scripts/generate_image.py:147

Python HTTP libraries

with urllib.request.urlopen(req, timeout=180) as response:

The urlopen call performs the external request to Gemini. Network transfer is core behavior and should be disclosed to users.

Medium

SKILL.md:12-18

Ruby/shell backtick execution

```bash

This fenced block instructs users or agents to run the local Python generator script. External command execution is expected but it can make network requests and write files.

Medium

SKILL.md:22-24

Ruby/shell backtick execution

```bash

This fenced usage block shows direct execution of scripts/generate_image.py. The command is intentional but still invokes local code with user-controlled arguments.

Medium

SKILL.md:44-47

Ruby/shell backtick execution

```bash

This fenced example runs the Python script with a reference image and output path. It invokes local code that may read files and send data to the API.

Medium

SKILL.md:51-54

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator with a reference image. The command is intentional but confirms external command use.

Medium

SKILL.md:58-61

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator with multiple reference images. That can read local images and send them to the API.

Medium

SKILL.md:69-76

Ruby/shell backtick execution

```bash

This fenced workflow block contains several invocations of the Python generator. It confirms intended local command execution for batch image generation.

Medium

SKILL.md:102-105

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator to create an app icon. It is legitimate usage but confirms command execution by the skill workflow.

Medium

SKILL.md:108-111

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator to create a banner. It invokes local script execution with an output path.

Medium

SKILL.md:114-117

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator for a knowledge card. It is an intended external command invocation.

Medium

SKILL.md:120-123

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator with a reference image for style transfer. It can read the referenced file and send it externally.

Medium

SKILL.md:126-129

Ruby/shell backtick execution

```bash

This fenced example runs the Python generator with a character reference image. It confirms intended command execution and external processing of local files.

Low

scripts/generate_image.py:23

Python HTTP libraries

import urllib.request

The imported urllib.request module is used later to call the Gemini API. This is expected for the feature but confirms external network capability.

Low

scripts/generate_image.py:34

Hardcoded URL

API_BASE = "https://generativelanguage.googleapis.com/v1beta/models"

The hardcoded endpoint targets the Google Generative Language API. It is legitimate for the skill, but it is still a fixed third-party destination.

Risk Factors

🌐 Network access (9)

README_CN.md:38 README_CN.md:54 README.md:38 README.md:54 scripts/generate_image.py:23 scripts/generate_image.py:144 scripts/generate_image.py:147 scripts/generate_image.py:34 scripts/generate_image.py:43

📁 Filesystem access (4)

README_CN.md:57 README_CN.md:57 README.md:57 README.md:57

🔑 Env variables (15)

README_CN.md:29 README_CN.md:32 README_CN.md:40 README_CN.md:43 README.md:29 README.md:32 README.md:40 README.md:43 scripts/generate_image.py:5 scripts/generate_image.py:37 scripts/generate_image.py:45 scripts/generate_image.py:101 scripts/generate_image.py:103 scripts/generate_image.py:105 scripts/generate_image.py:108

⚙️ External commands (39)

Audited by: codex View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/lxfater-nano-image-generator/audits/7?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/lxfater-nano-image-generator/security.svg)](https://skillstore.io/skills/lxfater-nano-image-generator?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/lxfater-nano-image-generator?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/lxfater-nano-image-generator/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/lxfater-nano-image-generator.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

lxfater. (2026). nano-image-generator security audit report (audit version 7) [Author version unspecified]. Skillstore. https://skillstore.io/skills/lxfater-nano-image-generator/audits/7

BibTeX citation

@techreport{lxfater-lxfater-nano-image-generator-2026,
  author = {lxfater},
  title = {nano-image-generator security audit report (audit version 7)},
  institution = {Skillstore},
  year = {2026},
  number = {7},
  url = {https://skillstore.io/skills/lxfater-nano-image-generator/audits/7},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "nano-image-generator security audit report (audit version 7)"
version: "unspecified"
type: report
authors:
  - name: "lxfater"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/lxfater-nano-image-generator/audits/7"
identifiers:
  - type: other
    value: "skillstore:lxfater-nano-image-generator:audit:7"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Generate Product Graphics

Create app icons, UI assets, and feature illustrations directly from project prompts.

Produce Campaign Images

Create banners, social images, and launch visuals with selected aspect ratios and sizes.

Maintain Visual Consistency

Use reference images to keep a consistent style or character across a sequence of assets.

Try These Prompts

Basic Image

Generate an image of a friendly robot mascot waving, with a clean modern style, and save it to the generated folder.

Sized Banner

Create a 16:9 website hero banner for a productivity app with abstract geometric shapes and a calm professional color palette.

Reference Style

Use the reference image as the visual style and generate a new mountain scene with matching colors, lighting, and texture.

Consistent Series

Generate a three-image knowledge card series with a consistent layout, matching typography space, and a clear visual theme.

Best Practices

Use specific prompts with subject, style, colors, composition, and intended use.
Store generated assets in project folders that match their purpose.
Review reference images before sending them to the external API.

Avoid

Do not hardcode real API keys in files committed to source control.
Do not send private reference images unless external processing is acceptable.
Do not rely on generated images for exact text, trademarks, or final brand approval.