Security audit
Security checks for vulnerabilities and agentic risk
This skill does what it claims: it sends a user-provided public product image URL to xAI/Grok to generate product descriptions.
Install only if you are comfortable sending public product image URLs and generated prompts to xAI using your XAI_API_KEY. Do not use it with private, internal, password-protected, or sensitive image links.
The skill advertises use of environment credentials and outbound network access but does not declare any explicit tool scope or permission boundaries. That creates an authorization ambiguity where an agent runtime may permit broader-than-expected access, increasing the chance of secret misuse or unreviewed external requests when the skill is invoked.
The skill is explicitly designed to transmit data to an external service at api.x.ai, which means user-provided content and prompts leave the local environment. In this context the external call is expected functionality, but it still presents a genuine security/privacy concern if users are not aware of third-party processing or if sensitive URLs are submitted.
# Requires: XAI_API_KEY environment variable
set -euo pipefail
API_URL="https://api.x.ai/v1/chat/completions"
_check_key() {
if [[ -z "${XAI_API_KEY:-}" ]]; then
The script sends a user-supplied image URL to xAI's external API as part of the request payload, but it does not clearly warn the user that the URL and associated image resource will be shared with a third party. This can create privacy and data-governance issues if users provide internal, sensitive, or personally identifying image links under the assumption processing is local.
The natural-language description states the skill 'Supports English, Chinese, Spanish output,' which constrains language handling to a fixed set of locales. This can be a locale-policy concern because the documentation presents a predetermined language set without explaining whether users can opt into other languages or why the limitation exists.
No suspicious patterns detected.