skills/itechmeat/llm-code/perplexity/Gen Agent Trust Hub

perplexity

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill consists of documentation and reference files (markdown). No executable scripts, persistence mechanisms, or privilege escalation patterns are present.\n- [EXTERNAL_DOWNLOADS]: The skill references standard libraries for the service, including the perplexityai Python package and the @perplexityai/perplexity npm package. These are external dependencies required to use the documented API.\n- [DATA_EXFILTRATION]: Authentication is handled via environment variables (e.g., PERPLEXITY_API_KEY), following security best practices to avoid hardcoded credentials.\n- [PROMPT_INJECTION]: No malicious instructions or bypass attempts were found. The skill includes specific advice on preventing model hallucinations by using citations instead of requesting URLs in prompts.\n- [COMMAND_EXECUTION]: Code snippets provided are for educational purposes and do not contain dangerous subprocess calls or unauthorized system commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 03:29 PM