perplexity-ai

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

• Third-party content exposure detected (high risk: 0.85). Perplexity AI actions (e.g., perplexityai_create_chat_completion / perplexityai_execute_agent) perform runtime web search and fetch/ground responses with citations, so outsider-authored web page text is ingested into the LLM context via Perplexity’s web-grounded completion pipeline.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires running "npx @useclawlink/cli" which fetches and executes remote code from the npm package (https://www.npmjs.com/package/@useclawlink/cli) at runtime to connect/control Perplexity, so a remote executable dependency is required and executed during runtime.