service-omni-queue-deploy
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data returned from Salesforce API queries (such as EntityDefinition and QueueSobject metadata). It mitigates potential injection risks by applying regular expression validation to DeveloperNames and XML escaping to metadata labels before the data is used in downstream deployment operations.
- [COMMAND_EXECUTION]: The bash script executes the
sfandjqCLI tools. To prevent command injection, it validates user-supplied arguments against strict patterns (e.g.,^[A-Za-z0-9_]{1,80}$) and enforces a hardcoded whitelist for thesobject_typeparameter. - [DYNAMIC_EXECUTION]: The skill dynamically generates Salesforce metadata XML files and project configuration in a temporary directory at runtime. This process is secured through the use of randomized paths and rigorous sanitization of all variables interpolated into the XML structure.
Audit Metadata