perf-ops

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to detect installed performance utilities and gather system baseline metrics like CPU, memory, and disk I/O usage. These operations are limited to standard system monitoring commands (e.g., top, free, iostat).
  • [EXTERNAL_DOWNLOADS]: The reference documentation provides installation instructions for well-known and trusted performance tools, including k6, clinic.js, and py-spy. The k6 installation process involves adding an official GPG key and repository from a well-known service provider (Grafana).
  • [REMOTE_CODE_EXECUTION]: While the skill orchestrates the execution of profiling tools, it does not involve running arbitrary remote scripts from untrusted sources. All profiling and benchmarking operations are performed using established, locally-installed tools or well-known packages from official registries.
  • [DATA_EXFILTRATION]: The skill reads profiling artifacts (e.g., .prof, .svg, .bin files) to interpret performance bottlenecks. No evidence of unauthorized data exfiltration or credential harvesting was detected.
  • [PROMPT_INJECTION]: The skill implements a multi-agent orchestration pattern with structured templates. It does not contain instructions aimed at bypassing AI safety guardrails or overriding system prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 11:30 AM