Security audit
Security checks for vulnerabilities and agentic risk
This skill appears to do what it claims: it sends LEGO build prompts and optional image URLs to an external Craftsman/DeepNLP API using a named API key, with no evidence of hidden persistence or unrelated access.
Install only if you are comfortable sending your LEGO prompt, optional image URLs, and the OneKey API key to the DeepNLP/Craftsman service. Avoid using confidential prompts, private image links, personal data, or regulated content unless that external service is approved for your use.
The skill requires environment access for an API key and performs outbound network calls, but it does not declare a restrictive tool scope such as explicit permissions or allowed-tools. This creates an under-scoped skill boundary where a host agent may grant broader capabilities than users expect, increasing the chance of unintended secret exposure or unreviewed external communication.
The skill instructs users to submit prompts and optionally images to an external API, but it does not clearly warn that this content leaves the local environment and is processed by a third-party service. Users may unknowingly transmit sensitive text or images, creating privacy, confidentiality, and compliance risks that are amplified because the skill is specifically designed to accept rich user-provided content.
The script transmits user-supplied prompts and image URLs to an external service, but it does not provide an explicit user-facing notice at the point of execution that potentially sensitive input will leave the local environment. In a skill context, users may assume processing is local, so this creates a privacy and data-handling risk, especially if prompts or image references contain confidential project details.
The script sends the user-supplied prompt and image URLs to a third-party endpoint without any explicit user-facing disclosure, confirmation, or consent step. Because prompts and referenced image URLs can contain sensitive project details, personal data, or internal links, this creates a privacy and data-handling risk even though the transmission uses HTTPS.
The code reads DEEPNLP_ONEKEY_ROUTER_ACCESS and uses it as an authentication header for the outbound request. Although missing-key handling exists, there is no user-facing warning or explanatory comment near execution flow that the script uses a sensitive credential to contact an external service.
Detected: suspicious.env_credential_access
