perplexity

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [NO_CODE] (SAFE): The skill is purely informational, providing Markdown documentation and guidance for tool selection. It does not include any scripts, binaries, or automated tasks.
  • [COMMAND_EXECUTION] (SAFE): There are no shell commands, subprocess calls, or system-level operations present in the files.
  • [DATA_EXFILTRATION] (SAFE): No sensitive file paths, credential patterns, or unauthorized network exfiltration behaviors were identified.
  • [PROMPT_INJECTION] (SAFE): The instructions are designed to constrain and guide the agent's behavior for efficiency and cost-effectiveness, rather than bypassing safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:56 PM