generate-image

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs asking the user for an API Key and shows command examples and parameters (e.g., --apikey ) that require embedding the secret verbatim in commands or config, which forces the LLM to handle/output secrets directly.

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly calls a user-configurable OpenAI-compatible image API (endpoint constructed from apiBaseUrl -> ${apiBaseUrl}/images/generations) and parses streaming JSON events and returned image URLs (see scripts/generate-image.mjs pickImage/readStreamResponse/saveGeneratedImage and the SKILL.md /gi workflow), meaning it ingests untrusted third-party responses and uses them to decide fallback behavior and to fetch/download arbitrary URLs, which could enable indirect prompt injection.