npm-package

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Agent skill" section (library package template) explicitly instructs the agent to fetch the latest README from a public URL (curl -s https://raw.githubusercontent.com/owner/repo/main/README.md), so the agent would ingest untrusted, user-controlled third-party content that can materially influence its behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL.md template explicitly instructs agents to fetch remote instructions at runtime via "curl -s https://raw.githubusercontent.com/owner/repo/main/README.md", meaning the skill relies on that external URL to provide prompt/context that directly controls agent behavior.