← Back to Skills Marketplace
Image Generate
by
zhangsiyuan123123
· GitHub ↗
· v1.0.0
2899
Downloads
0
Stars
153
Active Installs
1
Versions
Install in OpenClaw
/install image-generate
Description
使用内置 image_generate.py 脚本生成图片, 准备清晰具体的 `prompt`。
Usage Guidance
Use a scoped Ark/Volcengine API key, avoid sensitive personal data or secrets in prompts, install the required SDK only from a trusted source, and set IMAGE_DOWNLOAD_DIR only to a directory where you want generated images saved.
Capability Analysis
Type: OpenClaw Skill
Name: image-generate
Version: 1.0.0
The skill bundle is classified as suspicious due to a critical shell injection vulnerability and potential arbitrary file write. The `SKILL.md` instructs the agent to execute `python scripts/image_generate.py "<prompt>"`, directly embedding the user-provided prompt into a shell command without apparent sanitization, which could lead to arbitrary command execution if the agent's execution environment does not handle this. Additionally, the `scripts/image_generate.py` script uses the `IMAGE_DOWNLOAD_DIR` environment variable to determine where to save generated images, posing a risk of arbitrary file writes or path traversal if this variable can be controlled by an attacker.
Capability Assessment
Purpose & Capability
The purpose and capability are coherent: the skill runs an included Python script that calls the Volcengine Ark image-generation SDK with a user-provided prompt. There is a minor documentation mismatch: the skill says it returns image URLs, while the code downloads images and prints local file paths.
Instruction Scope
Instructions are scoped to image generation. The documented command includes a quoted prompt placeholder; agents should pass the prompt as one argument rather than interpolating untrusted text through a shell.
Install Mechanism
No installer or dependency setup is included, while the script imports volcenginesdkarkruntime. Users may need to install that SDK separately from a trusted source.
Credentials
Use of MODEL_IMAGE_API_KEY or ARK_API_KEY is expected for this provider integration. Prompts are sent to an external image-generation service, and this should be considered before entering sensitive content.
Persistence & Privilege
The script does not show background persistence, privilege escalation, broad local reads, credential exfiltration, or destructive behavior. It writes generated image files to IMAGE_DOWNLOAD_DIR or the current directory using generated filenames.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install image-generate - After installation, invoke the skill by name or use
/image-generate - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of image-generate skill:
- Generate images from clear and specific text prompts using the included image_generate.py script.
- Returns the image URL as output; prints errors on failure.
- Supports API key authentication via multiple environment variables.
- Documentation includes example usage and recommended steps.
Metadata
Frequently Asked Questions
What is Image Generate?
使用内置 image_generate.py 脚本生成图片, 准备清晰具体的 `prompt`。 It is an AI Agent Skill for Claude Code / OpenClaw, with 2899 downloads so far.
How do I install Image Generate?
Run "/install image-generate" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Image Generate free?
Yes, Image Generate is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Image Generate support?
Image Generate is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Image Generate?
It is built and maintained by zhangsiyuan123123 (@zhangsiyuan123123); the current version is v1.0.0.
More Skills