alterlab-infographics

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly performs automatic web research using Perplexity Sonar Pro (described in SKILL.md) and the scripts (scripts/generate_infographic_ai.py: research_topic and web_search) and then injects the returned, web-sourced content into the generation prompt via _enhance_prompt_with_research, so untrusted third‑party content is read and can materially influence generation and iterative tool actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill makes runtime calls to https://openrouter.ai/api/v1 (used for Perplexity Sonar research and Nano Banana/Gemini generation/review) and explicitly injects the returned research content into the generation prompt, so external responses can directly control agent instructions.