setup-server
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill provides highly structured instructions for server hardening, including mandatory SSH verification steps, firewall (UFW) configuration, and fail2ban setup to mitigate brute-force attacks.
- [SAFE]: Robust safety invariants are defined to prevent common operational risks, such as losing remote access, unintended data deletion, or the exposure of internal database ports to the public internet.
- [SAFE]: The instructions explicitly prohibit dangerous execution patterns, such as piping unknown remote scripts to a shell (curl | bash), and enforce the use of official software repositories for installations.
- [SAFE]: Secret management is handled safely by instructing the agent to use local environment files with restrictive permissions and forbidding the logging or reporting of sensitive credentials.
- [NO_CODE]: The skill consists entirely of markdown-based instructions for an AI agent and does not include any accompanying scripts or binary executables.
Audit Metadata