skills/salesforcecommercecloud/b2c-developer-tooling/b2c-business-manager-extensions/Gen Agent Trust Hub
b2c-business-manager-extensions
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes how to build administrative interfaces that ingest and render data from untrusted sources, which represents a surface for indirect prompt injection attacks.
- Ingestion points: The controller and template examples in
SKILL.mdandreferences/EXTENSIONS-XML.mddemonstrate accessing data fromrequest.httpParameterMap(e.g.,request.httpParameterMap.message.stringValue). - Boundary markers: There are no explicit boundary markers or "ignore instructions" delimiters shown in the templates that render this external data.
- Capability inventory: The skill demonstrates how to create controllers that can render ISML templates, perform redirects based on user input, and write to system logs via the
LoggerAPI. - Sanitization: The provided examples show direct rendering of request parameters into the HTML templates without explicit validation or escaping, which could allow maliciously crafted input to influence the agent's perception of the page context.
Audit Metadata