occ-headless-frontend
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process data from external SAP Commerce Cloud OCC APIs (e.g., product descriptions, CMS components, and search results). This represents a vulnerability surface where malicious content could be embedded in the backend data to influence agent behavior. The skill mitigates this risk by including a 'Hard rule' in SKILL.md that mandates sanitization of HTML from CMS and product descriptions before rendering. (1) Ingestion points: Data fetched from OCC REST APIs as described in SKILL.md, references/catalog-search.md, and references/cms.md. (2) Boundary markers: Explicit hard rules for HTML sanitization are provided in SKILL.md. (3) Capability inventory: The skill uses network fetching via fetch, Angular HttpClient, or Nuxt $fetch and implements dynamic component rendering. (4) Sanitization: Framework-specific sanitization techniques (e.g., Angular's native sanitizer, DOMPurify) are recommended throughout the reference documentation.
- [SAFE]: The skill provides technical instructions and code snippets that adhere to industry security standards for headless commerce, such as the use of httpOnly cookies, confidential OAuth clients for server-side operations, and secure PKCE flows for browser-based authentication.
Audit Metadata