linkedin-content

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Yes — both URLs host a remote shell-install endpoint (curl https://cli.inference.sh | sh) on an unverified domain, and piping an external .sh into a shell from a non‑well‑known/third‑party site is a high‑risk pattern that can execute arbitrary malware or backdoors.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md Quick Start explicitly instructs running a web search assistant (infsh app run tavily/search-assistant with a query for "LinkedIn viral post examples…") to research public LinkedIn/online content, meaning the agent would fetch and read untrusted, user-generated third‑party posts and use those findings to shape its output.

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

• Potentially malicious external URL detected (high risk: 0.90). The Quick Start instructs a runtime install via "curl -fsSL https://cli.inference.sh | sh", which fetches and immediately executes remote code from https://cli.inference.sh and is presented as a required dependency for the skill, so it directly controls runtime behavior.