skills/shinyorg/skills/shiny-vpn/Gen Agent Trust Hub

shiny-vpn

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill mentions the installation of system packages (network-manager-strongswan or NetworkManager-strongswan) on Linux via apt or dnf. This is a documented requirement for supporting IKEv2 VPNs through NetworkManager on that platform.
  • [CREDENTIALS_UNSAFE]: The skill demonstrates code for handling sensitive VPN credentials, including shared secrets, usernames/passwords, and client certificates. The documentation correctly notes that these secrets are stored in secure, platform-native locations such as the iOS/macOS Keychain or Android's provisioned profile storage.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for processing untrusted data to generate VPN configurations.
  • Ingestion points: Properties of the VpnProfile object, such as Name, ServerAddress, and LocalIdentifier, which are often derived from user-provided values.
  • Boundary markers: The instructions do not define specific delimiters for these fields.
  • Capability inventory: The skill uses SaveProfile to write to system configurations and Connect to initiate network operations.
  • Sanitization: The skill relies on the underlying library and operating system to validate and sanitize input strings during profile provisioning.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 12:35 AM