harden
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONNO_CODE
Full Analysis
- [NO_CODE] (SAFE): This skill consists entirely of a Markdown file (SKILL.md) and does not include any executable scripts, binaries, or package dependencies.
- [PROMPT_INJECTION] (SAFE): Indirect Prompt Injection Surface: 1. Ingestion points: Data retrieved from remote servers during audit and report commands. 2. Boundary markers: Absent in documentation. 3. Capability inventory: Remote administrative command execution via SSH. 4. Sanitization: Absent in documentation. The risk is rated SAFE as no implementation code is provided to be exploited.
- [COMMAND_EXECUTION] (SAFE): The skill describes remote server management and administrative command execution, which is its primary stated purpose for security hardening.
Audit Metadata