landing-page-design

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). These URLs host a remote install script (curl | sh) and direct binary distribution from a single, not widely-known domain — a risky pattern because executing a fetched shell script and installing unsigned/only-checksummed binaries can deliver malware unless the domain, script, and checksums are independently verified and trusted.

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md Quick Start and examples explicitly run third-party search/QA apps (e.g., "infsh app run tavily/search-assistant" to research competitor landing pages and "infsh app run exa/answer" for social-proof stats), which fetch and expose public web content that the agent would read and use to drive design decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The Quick Start explicitly runs a remote installer with "curl -fsSL https://cli.inference.sh | sh" (which then downloads binaries from dist.inference.sh), so these URLs are fetched and executed at runtime and are required for the skill's image-generation workflow.