case-management-setup
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill employs the
Bashtool to run standard Salesforce CLI (sf) commands. These queries are used to retrieve metadata and configuration details (such as assignment rules, queue memberships, and entitlement statuses) from a Salesforce environment to assist in troubleshooting. All commands use placeholders like<alias>and<userId>, ensuring that execution is scoped to user-defined targets. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill retrieves data from external Salesforce orgs via
sf data query(e.g., in the 'Case Assignment Automation' and 'Email-to-Case' sections ofSKILL.md). - Boundary markers: The instructions do not define specific delimiters to wrap the output of CLI commands before the agent processes them.
- Capability inventory: The skill has access to the
Bashtool for running CLI commands and theReadtool for context gathering. - Sanitization: No explicit sanitization or output filtering is defined for the data returned from the Salesforce API via the CLI, though the targeted fields (metadata names and statuses) are low-risk for injection.
Audit Metadata