Salesforce Architect
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The content consists entirely of informational text and documentation templates related to Salesforce architecture. It provides guidance on platform governance, scalability, and adherence to governor limits without including any executable logic.
- [CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or sensitive credentials were detected in the skill instructions or templates.
- [DATA_EXFILTRATION]: No network-enabled commands, unauthorized data transfer mechanisms, or sensitive file access patterns were identified.
- [PROMPT_INJECTION]: No attempts to bypass safety filters or override agent behavior were detected; the instructions are consistent with the professional persona described.
- [INDIRECT_PROMPT_INJECTION]: Although the skill mentions processing data from external sources like Data Cloud and Agentforce, it proactively recommends security mitigations such as using the Einstein Trust Layer for PII masking and topic classification.
Audit Metadata