skypilot-multi-cloud-orchestration

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an inherent attack surface by design, as it facilitates the execution of code and configuration (YAML) on cloud infrastructure. * Ingestion points: The skill describes processes for loading YAML task and service files, as well as syncing local project directories to remote clusters as documented in SKILL.md and advanced-usage.md. * Boundary markers: The skill uses structured YAML schemas to delimit task properties and resource definitions. * Capability inventory: The skill utilizes cloud provider APIs to provision compute resources, synchronizes local file systems to remote nodes, and executes shell commands on remote instances. * Sanitization: Documentation emphasizes the use of built-in secret management features and .skyignore files to prevent accidental exposure of sensitive local data during the synchronization process.
  • [SAFE]: The skill follows established security best practices for infrastructure management. It explicitly guides users to manage sensitive credentials like API keys and tokens using environment variable interpolation or dedicated secret management features rather than hardcoding them, as evidenced in the Environment Management and Cost Optimization sections.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 05:50 AM