agentforce-gpt-email-generation-configure
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill manages organizational settings by invoking internal platform endpoints through
PATCHandGETrequests to paths under/headless/invoke/platform/gptemail-generation/. These operations allow the agent to modify specific organizational preferences (e.g.,EinsteinGPTSalesEmails,CallSummarization,ConversationMining) and trigger background services like Conversation Mining refreshes. - [SAFE]: The instructions explicitly define the required user permissions for each administrative task, such as 'Modify Sales Email prompt access' and 'Sales Record Summaries access'. This ensures the agent acknowledges the authorization context required for these operations.
- [SAFE]: No indicators of malicious behavior, such as prompt injection, data exfiltration, or obfuscation, were found. The skill processes structured system state data and uses strictly typed inputs (Booleans) for configuration changes, minimizing potential attack surfaces.
Audit Metadata