pptx-posters

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text directly via the user-provided “prompt” (and its derivatives) into scripts/generate_schematic*.py → generate_schematic_ai.py, which sends that text to an external LLM for image generation and quality review.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's Python scripts call the OpenRouter API at runtime (https://openrouter.ai/api/v1) and use the model review outputs (Gemini) to iteratively modify generation prompts, so external responses directly control prompts/instructions.