service-itsm-agentic-setup-incident-sla-configure
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill demonstrates safe and professional design for administrative automation within a Salesforce environment.
- [DATA_EXFILTRATION]: The skill implements a robust privacy measure by explicitly forbidding the output of raw Salesforce record IDs (15/18-character strings) in user-facing text. This prevents the exposure of internal system identifiers and metadata to unauthorized users.
- [PROMPT_INJECTION]: No evidence was found of prompt injection attempts, role-play bypasses, or instructions to ignore safety guidelines.
- [REMOTE_CODE_EXECUTION]: The skill does not download or execute external scripts. All operations are performed via structured tool calls to a platform-hosted MCP server (
headless-360). - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface by reading picklist values and incident records from the org. However, the risk is minimal as the skill operates in a managed administrative context and includes instructions to verify and reconcile data before mutation.
Audit Metadata