applying-cms-brand
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill uses official tools and follows standard workflows for brand management within a Salesforce environment.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it is designed to follow instructions dynamically retrieved from the Salesforce CMS. * Ingestion points: The promptBody output from the get_brand_instructions tool in SKILL.md. * Boundary markers: The skill does not implement specific delimiters or 'ignore' warnings for the instructions contained within the brand data. * Capability inventory: The skill is restricted to content generation and does not utilize high-risk tools for file system access, network exfiltration, or shell execution. * Sanitization: There is no evidence of sanitization for the retrieved brand content, as it is treated as a trusted internal organizational resource.
Audit Metadata