service-omni-work-skill-routing-configure
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
sfCLI (Salesforce CLI) to interact with the target environment andjqfor JSON processing. These are standard, well-documented tools for Salesforce development and are used here for their intended purpose. - [COMMAND_EXECUTION]: A Python-based test suite (
scripts/tests/_bootstrap.py) usessubprocess.run()to invoke the bash scripts in an isolated, mock environment for contract testing. This is a common and safe practice for testing shell scripts. - [SAFE]: The skill implements a
safe_to_writeguard that queries the target organization's metadata (IsSandbox, TrialExpirationDate) to refuse operations against production environments, preventing accidental configuration changes. - [SAFE]: All user-supplied inputs ($RELATED_ENTITY, $FIELD, $SKILL, $VALUE, etc.) are validated against strict alphanumeric and character-set regular expressions before being passed to any shell command or embedded in XML, effectively mitigating command and XML injection risks.
- [SAFE]: The skill fetches configuration and queries existing metadata only from the authenticated Salesforce organization via official CLI tools. No external, untrusted network requests or remote code downloads were identified.
Audit Metadata