mcloud-auth

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to interact with the Medusa Cloud CLI tool (mcloud) to manage authentication states and environment context. All identified commands are scoped to official vendor functionality for managing accounts and project configurations.
  • [INDIRECT_PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by reading and parsing JSON output from the Medusa Cloud service (e.g., resolving organization IDs and project handles). However, the instructions utilize standard best practices, such as shell variable quoting ("$ORGANIZATION_ID"), to safely handle data returned from the CLI tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 02:10 PM