linkedin-content

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). inference.sh is a third‑party .sh domain referenced as the source for a CLI/installer (runnable code), and because it’s not an obviously established vendor or a trusted package manager and could serve scripts/binaries, it represents a moderate-to-high risk unless independently verified.

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 explicitly instructs running infsh app run tavily/search-assistant with queries like "LinkedIn viral post examples 2024…" which fetches and ingests public, user-generated web content that the agent is expected to read and use to shape posting behavior, so untrusted third-party content could materially influence its actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires and uses the inference.sh CLI (https://inference.sh) at runtime via "infsh app run ..." commands (e.g., tavily/search-assistant, falai/flux-dev-lora), which execute remote apps/code and return content the skill depends on, so this external URL is a runtime dependency that executes remote code.