pplx-cli

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

• Third-party content exposure detected (low risk: 0.10). SKILL.md describes pplx search web and pplx content snippets fetching and extracting query-relevant page excerpts/snippets from the live web (user-supplied queries and URLs), so outsider-authored free text can be ingested into the LLM via returned snippets/text.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runs remote code at runtime via a piped install script (curl -fsSL https://github.com/perplexityai/perplexity-cli/releases/latest/download/install.sh | sh), which fetches and executes whatever that URL serves.