sf-ai-agentforce-conversationdesign
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of documentation, templates, and a local Python script designed to assist in the creation and validation of Salesforce Agentforce conversation designs. No malicious patterns, data exfiltration, or unauthorized command executions were found.
- [COMMAND_EXECUTION]: The skill includes a local Python script
hooks/scripts/instruction-quality-validator.pywhich is executed via a lifecycle hook defined in.claude/hooks.yaml. This script performs local regex-based analysis of text to validate instruction quality and does not perform any network operations or sensitive file access. - [INDIRECT_PROMPT_INJECTION]: While the skill defines patterns for handling untrusted user input, it explicitly provides a 'Guardrail Hierarchy' guide (references/guardrail-hierarchy.md) that describes multi-layer safety models including the Einstein Trust Layer, Topic Classification, and deterministic Flow/Apex logic to mitigate such risks.
Audit Metadata