napkin-math

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to reference the sirupsen/napkin-math GitHub repository for hardware and network performance data. This is a legitimate reference for the skill's purpose and does not involve executing untrusted scripts.
  • [DATA_EXFILTRATION]: No patterns for accessing sensitive files (e.g., credentials, SSH keys, environment variables) or exfiltrating data to external servers were found.
  • [COMMAND_EXECUTION]: The skill does not contain any instructions to execute shell commands, subprocesses, or system-level scripts.
  • [PROMPT_INJECTION]: The instructions focus entirely on mathematical estimation and do not attempt to override agent safety protocols or system guidelines.
  • [REMOTE_CODE_EXECUTION]: There are no indicators of remote code execution, package installations, or dynamic code evaluation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 03:39 PM