customer-persona

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

• Third-party content exposure detected (low risk: 0.10). The required workflow uses search and research tools (tavily/search-assistant, exa/search, exa/answer) to gather target market demographics and pain points from web search results, which is classified as low risk since the agent actively fetches external content.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs fetching and executing a remote install script (raw.githubusercontent.com/inference-sh/skills/.../cli-install.md) and installing the "belt-sh/cli" skill package at runtime, both from an unverified third‑party GitHub-hosted source without official vendor authentication, making this a common/reputable unofficial operational dependency.