imagegen

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). Contains an automated post-install installer that clones a remote repo into the user’s home and installer scripts that copy configuration and will send a locally-stored API key to a configured endpoint — a supply-chain/post-install script risk and potential credential-exfiltration vector if the configured endpoint is malicious.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow ingests outsider-authored free text directly as the user’s image generation/edit prompts (and optional verbatim text) via step 2/4 prompt augmentation, then sends that text to either the built-in image_generation_call or the custom endpoint.