Skills
skills/membranedev/application-skills/cloudback/Gen Agent Trust Hub

cloudback

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Installs the @membranehq/cli tool from the official npm registry to facilitate communication with the Cloudback connector.
  • [COMMAND_EXECUTION]: Utilizes shell-based interactions with the membrane CLI to perform authentication, connection setup, and execution of Cloudback actions.
  • [PROMPT_INJECTION]: Contains a potential surface for indirect prompt injection as it processes external action schemas and descriptions (SKILL.md). However, no boundary markers are bypassable and the capability is limited to vendor-authorized actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 07:59 AM
Security Audit — agent-trust-hub — cloudback