product-hunt-launch

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

• Third-party content exposure detected (high risk: 0.85). The skill’s workflow uses the inference.sh web-search apps (e.g., tavily/search-assistant / exa/search) to fetch public web pages at runtime and feed their extracted text into the agent’s LLM context, which is outsider-authored free text via the web-search tool path.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill invokes remote apps/models at runtime via the inference.sh CLI (e.g., "infsh app run falai/flux-dev-lora" and "infsh app run tavily/search-assistant"), which execute remote code on https://inference.sh and are required for image generation and research, so this is a runtime external dependency that executes remote code.