experience-cms-brand-apply
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements standard brand management workflows within the Salesforce ecosystem using platform-specific tools (
search_brands,get_brand_instructions). - [PROMPT_INJECTION]: The skill retrieves brand instructions from Salesforce CMS and incorporates them into the agent's context, creating an indirect prompt injection surface.
- Ingestion points: Brand instructions are fetched via the
get_brand_instructionstool, specifically from thepromptBodyfield. - Boundary markers: The agent is instructed to follow rules within the retrieved
promptBodywithout explicit isolation delimiters. - Capability inventory: Access is restricted to Salesforce CMS search and retrieval tools; no arbitrary command execution or external network operations are present in the skill code.
- Sanitization: The skill relies on the integrity of the data source (Salesforce CMS) and does not specify additional client-side sanitization.
Audit Metadata