platform-sandbox-configure
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests sandbox-related data which serves as a potential vector for indirect prompt injection, though its design minimizes risk. \n
- Ingestion points: The skill retrieves sandbox names and statuses from API responses (SKILL.md). \n
- Boundary markers: Mandatory user confirmation is required before any destructive actions, such as deletion or activation. \n
- Capability inventory: The skill has the ability to delete, create, and refresh sandboxes using the Salesforce CLI. \n
- Sanitization: It implements pre-condition checks, ensuring the sandbox is in the correct state before proceeding with lifecycle changes.
Audit Metadata