perplexity-search
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the 'litellm' Python package, which is a standard library for interacting with multiple LLM providers. Installation instructions correctly use standard package managers like 'uv' or 'pip'.
- [DATA_EXFILTRATION]: Network operations are directed to OpenRouter's official API endpoints (openrouter.ai) to facilitate searches. No unauthorized data transmission or exfiltration to unknown third-party domains was detected.
- [CREDENTIALS_UNSAFE]: The skill uses environment variables and '.env' files for API key management, which is a standard and secure practice. No hardcoded API keys or secrets were found in the scripts or documentation; only placeholders are used for demonstration.
- [COMMAND_EXECUTION]: The provided Python scripts perform standard argument parsing and API requests. There is no evidence of arbitrary command execution, shell injection, or unsafe subprocess spawning.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted data from web search results, it merely displays or saves this content. It does not pipe the output into an execution environment (like a shell or eval), effectively mitigating the risk of indirect prompt injection attacks.
Audit Metadata