perplexity

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill specifies dependencies on ai-agent-marketplace (PyPI) and @aiagenta2z/onekey-gateway (NPM). These are official packages associated with the skill's author (aiagenta2z) and are used for routing requests to AI models.
  • [CREDENTIALS_UNSAFE]: The Python scripts in the scripts/ directory include a hardcoded fallback string BETA_TEST_KEY_MARCH_2026. This is explicitly labeled as a demo/test key used only when the primary environment variable DEEPNLP_ONEKEY_ROUTER_ACCESS is missing.
  • [COMMAND_EXECUTION]: The skill provides Python scripts and CLI commands via npx to interact with the Perplexity API. These operations are restricted to the intended purpose of querying the AI models.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 02:40 PM
Security Audit — agent-trust-hub — perplexity