Security audit
Security checks for vulnerabilities and agentic risk
The skill’s image-generation purpose is coherent, but it needs review because API responses can choose unvalidated upload and download network destinations.
Install only if you are comfortable sending prompts, selected reference images, and an AI Hive API key to this service; use a virtual environment, avoid sensitive internal images, and consider running it from a network where unintended requests to private services would have limited impact.
scripts/imagegen.py:231Unvalidated API-Supplied Upload and Download URLs Enable Network Request Redirection
SKILL.md:53Unpinned Third-Party Dependency Installation Instruction
The skill invokes local scripts, installs dependencies, reads input images, may write outputs, and sends data to a remote API, but it does not declare any explicit tool or permission scope. This creates a trust and review gap: an agent may grant broader shell, file, environment, and network capabilities than users expect, increasing the risk of unintended file access or exfiltration when the skill is executed.
The manifest description and primary instructions are written in Chinese and present the skill behavior in that language without stating that users may choose another language or locale. This can violate a language/locale policy when a skill implicitly enforces one language without opt-in.
This code file contains user-facing natural-language strings and documentation entirely in Chinese, including operational instructions and prompts. Under the policy rule, forcing a specific language without opt-in is a natural-language policy violation unless the locale constraint is explicitly documented and justified, which is not present here.
The init flow opens a browser to a fixed service URL and appends skill-identifying query parameters (from=cli-skill&skill=...) that are not required to obtain an API key. This creates unnecessary telemetry about which skill a user is configuring and can leak usage metadata to the service or intermediaries, even though it does not expose the API key itself.
No suspicious patterns detected.