service-omni-service-channel-configure
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The script
scripts/deploy-and-report.shexecutessf(Salesforce CLI) andjqcommands to query org metadata and perform deployments. These are standard tools for Salesforce development and their usage is consistent with the skill's stated purpose. - [INDIRECT_PROMPT_INJECTION]: The skill processes data queried from the Salesforce org (via
EntityDefinitionandServiceChannel). It mitigates potential injection risks by validating theSOBJECT_TYPEinput against a strict whitelist in a Bashcasestatement before using it in queries or deployment logic. - [PRIVILEGE_ESCALATION]: The skill implements a
safe_to_writeguard that verifies if the target Salesforce org is a Sandbox, Developer Edition, or Trial org. It explicitly blocks execution on production environments to prevent unintended modifications. - [DATA_EXPOSURE]: While the skill queries organization details and metadata, it does so to determine deployment state. There is no evidence of sensitive data (like credentials) being hardcoded or exfiltrated to external domains.
Audit Metadata