find-skills

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs using npx skills find [query] and links to https://skills.sh/ and npx skills add <owner/repo@skill> (from GitHub or other sources), which fetches and surfaces open, user-contributed third-party skill metadata that the agent reads and can influence install/actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs running npx skills add owner/repo@skill and links to https://skills.sh/ (e.g. https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practices), which at runtime fetches and installs remote skill packages from external repositories that execute remote code and can modify/drive agent behavior.