@tank/character-creator

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill generates a canonical image via OpenAI's image endpoint (api.openai.com/v1/images/generations), captures the returned image URL, and then passes that external image_url to fal.ai's Flux Kontext endpoint (fal-ai/flux-pro/kontext) as a required runtime reference to drive subsequent prompt-guided generations, so external content fetched at runtime directly controls the agent's outputs.