perplexity-search
Fail
Audited by Snyk on Feb 16, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 0.80). The prompt includes explicit examples that embed an OpenRouter API key verbatim (both as an env var placeholder and—importantly—as a command-line argument like
--api-key sk-or-v1-your-key-here), which is an insecure pattern that would require or encourage the LLM to handle/output secret values directly.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The skill explicitly performs real-time web searches using Perplexity models via OpenRouter (e.g., "Perplexity provides real-time, web-grounded answers" and links to openrouter.ai/perplexity), ingesting and interpreting content from public websites and literature as part of its workflow, which exposes the agent to untrusted third-party content.
Audit Metadata