ai-content-pipeline

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). Although most links are documentation and media on a single third‑party domain (inference.sh), the prompt explicitly instructs piping a remote installer (curl https://cli.inference.sh | sh), includes a vague external audio URL (https://your-music.mp3) and a personal file‑hosting URL — all of which are common vectors for distributing malware if the source is not trusted.

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md explicitly includes a "Research" step using tavily/search-assistant and workflows like "Summarize this blog post" which show the agent ingesting and acting on public web/blog content (untrusted third‑party sources) as part of the pipeline, allowing that content to influence subsequent tool usage and outputs.

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

• Potentially malicious external URL detected (high risk: 0.90). The Quick Start instructs running "curl -fsSL https://cli.inference.sh | sh" which downloads and immediately executes remote code from https://cli.inference.sh and is required to install the infsh CLI used by the skill, so this is a high-risk runtime external dependency.