hz-simpleperf-debug

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The instructions utilize npx -y metavr to execute the profiling utility. This pattern involves fetching the package from the npm registry at runtime. Given that metavr is a documented tool from the skill's author (meta-quest), this is categorized as intended functionality for the skill's purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze performance data produced by the metavr CLI. \n
  • Ingestion points: Command output from metavr perf simpleperf subcommands in SKILL.md and reference files. \n
  • Boundary markers: The skill does not define specific delimiters for tool output. \n
  • Capability inventory: The skill uses the Bash tool to execute profiling commands and read system state. \n
  • Sanitization: No explicit sanitization is applied to the performance metrics before analysis, though the technical nature of hardware performance counters naturally limits injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 06:57 PM
Security Audit — agent-trust-hub — hz-simpleperf-debug