perplexity
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains technical documentation and code examples for the Perplexity API. All instructions are consistent with the intended purpose of building search-powered AI applications.- [SAFE]: The recommended dependencies, specifically the 'perplexityai' Python package and '@perplexityai/perplexity' Node.js package, are legitimate libraries associated with the service.- [SAFE]: Authentication instructions promote secure practices by advising the use of environment variables for API keys rather than hardcoding them.- [SAFE]: Advanced features such as remote browser sessions and multi-step Pro Search are documented as built-in service capabilities, with no evidence of deceptive or malicious behavior.- [SAFE]: While the skill interacts with external web content (Indirect Prompt Injection surface), it includes guidance to mitigate common risks, such as warning against hallucinated URLs in prompts and recommending the use of official citation fields.
Audit Metadata