creating-b2b-commerce-store
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Salesforce CLI (
sf) to perform metadata operations. It executessf org list metadatato discover digital experience configurations andsf project retrieve startto fetch theDigitalExperienceBundle. These are standard developer tools and workflows for the Salesforce platform. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection (Category 8) because it processes metadata names retrieved from a Salesforce organization. 1. Ingestion points: Output from the
sf org list metadatacommand in Step 4 ofSKILL.md. 2. Boundary markers: Absent; the instructions do not include specific delimiters or directives to ignore instructions that might be contained within the metadata names. 3. Capability inventory: The skill executes shell commands via the Salesforce CLI (SKILL.md). 4. Sanitization: No validation or sanitization logic is specified for the metadata names retrieved from the environment before they are displayed or used. - [COMMAND_EXECUTION]: The skill generates shell commands dynamically by interpolating user-provided input (the selected site name) into the
sf project retrievecommand in Step 6. While standard for CLI-based workflows, this represents a point where external strings enter the shell execution context.
Audit Metadata