perplexity-search
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implementation strictly follows its stated purpose of providing a search interface. No hidden malicious logic, obfuscation, or unauthorized access patterns were detected.
- [EXTERNAL_DOWNLOADS]: The skill relies on LiteLLM, a reputable and widely-used library for LLM API integration. It also references OpenRouter, a standard and well-known service provider in the AI ecosystem.
- [CREDENTIALS_UNSAFE]: The skill demonstrates safe credential management by utilizing environment variables and .env files for API keys, avoiding hardcoded secrets and providing a setup script that handles sensitive data locally and securely.
- [DATA_EXFILTRATION]: Network communication is limited to the official OpenRouter API endpoints required for the search functionality. No attempts to exfiltrate sensitive local data or communicate with unknown third-party domains were found.
- [COMMAND_EXECUTION]: The included Python scripts perform expected tasks such as environment validation and API calls. They do not execute arbitrary shell commands or process untrusted input in a way that would allow for command injection.
Audit Metadata