service-digital-engagement-channel-configure
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various
sf(Salesforce CLI) commands to interact with the target organization, including querying for existing queues and flows, deploying metadata files, and performing REST API PATCH requests to bind service agents. These commands are essential for its documented purpose of setting up messaging channels. - [DYNAMIC_EXECUTION]: Python 3 is used for inline JSON parsing and version comparison within shell scripts (e.g.,
check-api-version.shandSKILL.mdstep 15a). These scripts process output from the Salesforce CLI and do not download or execute external code. - [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided strings for channel labels, routing targets, and queue names. It includes a normalization script (
normalize-channel-name.sh) to sanitize labels into API-compliant strings usingsed, which mitigates potential injection risks in the primary metadata file name.
Audit Metadata