ubuntu-landscape

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the vendor-provided Membrane CLI using npm install -g @membranehq/cli@latest. This tool is the primary interface for the skill's functionality.
  • [COMMAND_EXECUTION]: Employs the membrane command-line utility to manage sessions, create connections to Ubuntu Landscape, and execute actions.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes external data from action executions.
  • Ingestion points: Data enters the context via the output of membrane action run.
  • Boundary markers: None identified in the provided instructions.
  • Capability inventory: Subprocess calls are made via the membrane CLI across commands like login, connect, and action run.
  • Sanitization: No explicit sanitization or validation of action output is described before processing by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 05:38 AM
Security Audit — agent-trust-hub — ubuntu-landscape