Skills baoyu-image-gen
📦
baoyu-image-gen
Content revision r1 High Risk ⚡ Contains scripts⚙️ External commands📁 Filesystem access🔑 Env variables🌐 Network access
Generate Images with OpenAI and Google
Image prompts often need provider setup, sizes, and repeatable output paths. This skill runs an API-based generator with aspect ratio, quality, and reference image options.
Supports: Claude Codex Code(CC)
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 "baoyu-image-gen" from https://skillstore.io/skills/jimliu-baoyu-image-gen.md and its manifest at https://skillstore.io/api/skills/jimliu-baoyu-image-gen/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.
Test it
Using "baoyu-image-gen". Prompt: A calm product background for a smartwatch, 16:9.
Expected outcome:
A saved wide image path and a short confirmation naming the provider and model.
Using "baoyu-image-gen". Prompt files: brand brief and campaign theme. Output: social-card.png.
Expected outcome:
A generated image file that combines both prompt files into one visual request.
Using "baoyu-image-gen". Reference image: source.png. Prompt: Make the object blue.
Expected outcome:
A saved variation that uses the reference image when a supported Google model is selected.
Security Audit
High Riskv4 • 7/6/2026 Open versioned report
The code mostly implements the documented image-generation workflow, and many static hits are documentation backticks or JavaScript template literals. Confirmed risks center on loading project and home env files, sending API credentials over network requests, and allowing project configuration to redirect provider base URLs.
5
Files scanned
860
Lines analyzed
20
Review items
0
False positives ignored
Confirmed security concerns (10)
High
Environment file access
const homeEnv = await loadEnvFile(path.join(home, ".baoyu-skills", ".env"));
The script reads home and project .baoyu-skills/.env files that may contain credentials. Project-local env files are untrusted in shared repositories.
High
Environment file access
const cwdEnv = await loadEnvFile(path.join(cwd, ".baoyu-skills", ".env"));
The script reads home and project .baoyu-skills/.env files that may contain credentials. Project-local env files are untrusted in shared repositories.
High
Environment file access
if (!process.env[k]) process.env[k] = v;
The script merges parsed env-file values into process.env. That lets project-local configuration affect later credential and endpoint handling.
High
Environment file access
if (!process.env[k]) process.env[k] = v;
The script merges parsed env-file values into process.env. That lets project-local configuration affect later credential and endpoint handling.
High
Environment file access
return process.env.GOOGLE_API_KEY || process.env.GEMINI_API_KEY || null;
The provider reads GOOGLE_API_KEY or GEMINI_API_KEY for outbound API requests. This is necessary for generation but is sensitive credential handling.
High
Environment file access
const base = process.env.GOOGLE_BASE_URL || "https://generativelanguage.googleapis.com";
GOOGLE_BASE_URL controls the outbound API endpoint. If supplied by a project env file, Google credentials and prompts can be sent to a non-Google host.
High
Environment file access
const baseURL = process.env.OPENAI_BASE_URL || "https://api.openai.com/v1";
OPENAI_BASE_URL controls the outbound API endpoint. If supplied by a project env file, OpenAI credentials and prompts can be sent to a non-OpenAI host.
High
Environment file access
const apiKey = process.env.OPENAI_API_KEY;
The provider reads OPENAI_API_KEY and uses it in an Authorization header for image generation. This is required but exposes credentials to the configured endpoint.
Show all 10 confirmed findings
High
Project Configuration Can Redirect API Credentials
The script loads project .baoyu-skills/.env values, then uses OPENAI_BASE_URL or GOOGLE_BASE_URL for requests that include API keys. A malicious project can set a custom endpoint and receive provider credentials.
The code explicitly loads cwd .baoyu-skills/.env into process.env and provider code uses base URL env vars for credentialed fetch calls.
Medium
Unpinned Runtime Invocation Through npx
The usage instructions invoke npx -y bun without a pinned package version. This can execute package-manager-resolved code before the local skill script runs.
The documented command path is clear, but the practical risk depends on how the host environment resolves npx and bun.
Capability review items (20)
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
Environment variable access (dot notation)
return process.env.GOOGLE_API_KEY || process.env.GEMINI_API_KEY || null;
The provider reads GOOGLE_API_KEY or GEMINI_API_KEY for outbound API requests. This is necessary for generation but is sensitive credential handling.
High
Environment variable object
return process.env.GOOGLE_API_KEY || process.env.GEMINI_API_KEY || null;
The provider reads GOOGLE_API_KEY or GEMINI_API_KEY for outbound API requests. This is necessary for generation but is sensitive credential handling.
High
Generic API/secret keys
return process.env.GOOGLE_API_KEY || process.env.GEMINI_API_KEY || null;
The provider reads GOOGLE_API_KEY or GEMINI_API_KEY for outbound API requests. This is necessary for generation but is sensitive credential handling.
High
Environment variable access (dot notation)
const apiKey = process.env.OPENAI_API_KEY;
The provider reads OPENAI_API_KEY and uses it in an Authorization header for image generation. This is required but exposes credentials to the configured endpoint.
High
Environment variable object
const apiKey = process.env.OPENAI_API_KEY;
The provider reads OPENAI_API_KEY and uses it in an Authorization header for image generation. This is required but exposes credentials to the configured endpoint.
High
Generic API/secret keys
const apiKey = process.env.OPENAI_API_KEY;
The provider reads OPENAI_API_KEY and uses it in an Authorization header for image generation. This is required but exposes credentials to the configured endpoint.
Medium
Environment variable access (bracket notation)
if (!process.env[k]) process.env[k] = v;
Values from .baoyu-skills/.env are copied into process.env when unset. Project-local configuration can influence provider endpoints and credentials used later.
Medium
Environment variable access (bracket notation)
if (!process.env[k]) process.env[k] = v;
Values from .baoyu-skills/.env are copied into process.env when unset. Project-local configuration can influence provider endpoints and credentials used later.
Medium
Environment variable object
if (!process.env[k]) process.env[k] = v;
Values from .baoyu-skills/.env are copied into process.env when unset. Project-local configuration can influence provider endpoints and credentials used later.
Medium
Environment variable object
if (!process.env[k]) process.env[k] = v;
Values from .baoyu-skills/.env are copied into process.env when unset. Project-local configuration can influence provider endpoints and credentials used later.
Medium
Fetch API call
const res = await fetch(buildGoogleUrl(pathname), {
This POST sends prompts, optional reference image data, and x-goog-api-key to the configured Google endpoint. The network egress is required but sensitive.
Medium
Hardcoded URL
const base = process.env.GOOGLE_BASE_URL || "https://generativelanguage.googleapis.com";
GOOGLE_BASE_URL controls the outbound API endpoint. If supplied by a project env file, Google credentials and prompts can be sent to a non-Google host.
Medium
Environment variable access (dot notation)
const base = process.env.GOOGLE_BASE_URL || "https://generativelanguage.googleapis.com";
GOOGLE_BASE_URL controls the outbound API endpoint. If supplied by a project env file, Google credentials and prompts can be sent to a non-Google host.
Medium
Environment variable object
const base = process.env.GOOGLE_BASE_URL || "https://generativelanguage.googleapis.com";
GOOGLE_BASE_URL controls the outbound API endpoint. If supplied by a project env file, Google credentials and prompts can be sent to a non-Google host.
Medium
Fetch API call
const res = await fetch(`${baseURL}/images/generations`, {
This POST sends the prompt and OpenAI bearer token to the configured endpoint. The network egress is required but sensitive.
Medium
Hardcoded URL
const baseURL = process.env.OPENAI_BASE_URL || "https://api.openai.com/v1";
OPENAI_BASE_URL controls the outbound API endpoint. If supplied by a project env file, OpenAI credentials and prompts can be sent to a non-OpenAI host.
Medium
Environment variable access (dot notation)
const baseURL = process.env.OPENAI_BASE_URL || "https://api.openai.com/v1";
OPENAI_BASE_URL controls the outbound API endpoint. If supplied by a project env file, OpenAI credentials and prompts can be sent to a non-OpenAI host.
Medium
Environment variable object
const baseURL = process.env.OPENAI_BASE_URL || "https://api.openai.com/v1";
OPENAI_BASE_URL controls the outbound API endpoint. If supplied by a project env file, OpenAI credentials and prompts can be sent to a non-OpenAI host.
Medium
Ruby/shell backtick execution
```bash
The usage block instructs agents to run npx -y bun, which can execute package-manager code before the local script. This is expected for the skill but remains an external command supply-chain risk.
Low
Fetch API call
const imgRes = await fetch(img.url);
The script fetches an image URL returned by the provider. With a custom endpoint, that response can trigger an extra outbound request.
Risk Factors
⚡ Contains scripts (2)
⚙️ External commands (54)
scripts/main.ts:8-37 scripts/main.ts:86 scripts/main.ts:108 scripts/main.ts:115 scripts/main.ts:136 scripts/main.ts:143 scripts/main.ts:150 scripts/main.ts:160 scripts/main.ts:165 scripts/main.ts:238 scripts/providers/google.ts:43 scripts/providers/google.ts:44 scripts/providers/google.ts:49 scripts/providers/google.ts:67 scripts/providers/google.ts:76 scripts/providers/google.ts:86 scripts/providers/google.ts:151 scripts/providers/google.ts:197 scripts/providers/openai.ts:85 scripts/providers/openai.ts:89 scripts/providers/openai.ts:96 SKILL.md:13 SKILL.md:14 SKILL.md:20-26 SKILL.md:26-48 SKILL.md:48-66 SKILL.md:66-72 SKILL.md:72 SKILL.md:73-74 SKILL.md:74-75 SKILL.md:75-76 SKILL.md:76 SKILL.md:77 SKILL.md:78 SKILL.md:79-80 SKILL.md:80-81 SKILL.md:81-82 SKILL.md:82-83 SKILL.md:83-89 SKILL.md:89-90 SKILL.md:90-91 SKILL.md:91-92 SKILL.md:92-93 SKILL.md:93-94 SKILL.md:94-96 SKILL.md:96 SKILL.md:100-108 SKILL.md:108-109 SKILL.md:109-111 SKILL.md:111-115 SKILL.md:115 SKILL.md:117-118 SKILL.md:118-132 SKILL.md:132-136
📁 Filesystem access (10)
🔑 Env variables (33)
scripts/main.ts:208 scripts/main.ts:211 scripts/main.ts:244 scripts/main.ts:245 scripts/main.ts:37 scripts/main.ts:208 scripts/main.ts:211 scripts/main.ts:244 scripts/main.ts:245 scripts/main.ts:29 scripts/main.ts:30 scripts/main.ts:31 scripts/main.ts:244 scripts/main.ts:245 scripts/main.ts:252 scripts/providers/google.ts:9 scripts/providers/google.ts:27 scripts/providers/google.ts:36 scripts/providers/google.ts:9 scripts/providers/google.ts:27 scripts/providers/google.ts:36 scripts/providers/google.ts:27 scripts/providers/google.ts:54 scripts/providers/openai.ts:4 scripts/providers/openai.ts:64 scripts/providers/openai.ts:65 scripts/providers/openai.ts:4 scripts/providers/openai.ts:64 scripts/providers/openai.ts:65 scripts/providers/openai.ts:65 scripts/providers/openai.ts:67 SKILL.md:89 SKILL.md:90
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.
Copy report link
https://skillstore.io/skills/jimliu-baoyu-image-gen/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/jimliu-baoyu-image-gen?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/jimliu-baoyu-image-gen?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/jimliu-baoyu-image-gen/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/jimliu-baoyu-image-gen.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
jimliu. (2026). baoyu-image-gen security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/jimliu-baoyu-image-gen/audits/4BibTeX citation
@techreport{jimliu-jimliu-baoyu-image-gen-2026,
author = {jimliu},
title = {baoyu-image-gen security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/jimliu-baoyu-image-gen/audits/4},
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: "baoyu-image-gen security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
- name: "jimliu"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/jimliu-baoyu-image-gen/audits/4"
identifiers:
- type: other
value: "skillstore:jimliu-baoyu-image-gen:audit:4"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: High 45
Architecture
85
Maintainability
85
Content
70
Community
83
Spec Compliance
What You Can Build
Create Design Drafts
Generate concept art, product scenes, or campaign visuals from short prompts.
Prepare Marketing Assets
Create wide, square, or portrait images for posts, slides, and landing pages.
Automate Visual Workflows
Add API image generation to agent workflows with chosen providers and output paths.
Try These Prompts
Create a Single Image
Use baoyu-image-gen to create one image of a modern desk setup at sunrise. Save it as desk-scene.png.
Set Shape and Quality
Generate a 16:9 hero image for a productivity app. Use 2k quality and save it in the project assets folder.
Use a Reference Image
Use the provided reference image and generate a blue variant with the same composition. Use the Google provider.
Run a Provider-Specific Set
Generate three options for a launch banner with OpenAI. Use a landscape aspect ratio and return concise result details.
Best Practices
- Keep API keys in user-level environment config and avoid committing project env files.
- Use explicit output paths inside the current project.
- Review custom base URLs before sending prompts or reference images.
Avoid
- Do not use secret documents as prompt files or reference images.
- Do not enable custom provider endpoints from untrusted projects.
- Do not request large batches without checking API cost and rate limits.
Frequently Asked Questions
Which providers does this skill support?
It supports OpenAI image generation and Google image models through API calls.
Does it need API keys?
Yes. Set OPENAI_API_KEY, GOOGLE_API_KEY, or GEMINI_API_KEY before running generation.
Can it use reference images?
Yes, but reference images are only sent to supported Google multimodal models.
Can I control image shape?
Yes. Use aspect ratio, size, quality, or Google image size options.
Where are images saved?
Images are written to the output path supplied with the image option.
Is custom endpoint support safe?
Only use custom base URLs that you trust because API credentials are sent to them.
Developer Details
Author
jimliuLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
a06681402992ceae98ba04d54cfd4ab004862696
Maintenance freshness
7/18/2026
Usage
21 downloads · 175 views