seven-powers

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • NO_CODE (SAFE): The skill consists entirely of markdown documentation. No Python scripts, JavaScript files, shell scripts, or package manager configuration files (e.g., package.json, requirements.txt) were detected across the 16 files.
  • PROMPT_INJECTION (SAFE): The instructional content contains no attempts to override system prompts, bypass safety filters, or extract system instructions. Examples provided are clearly delineated as strategy use cases.
  • DATA_EXFILTRATION (SAFE): No commands or patterns associated with accessing sensitive files (e.g., ~/.ssh, .env) or performing network requests (e.g., curl, wget) were found.
  • COMMAND_EXECUTION (SAFE): No subprocess calls, shell execution patterns, or privilege escalation attempts (e.g., sudo) were identified.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 01:27 AM