klaviyo-core-workflow-b
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes the official
klaviyo-apiSDK to interact with Klaviyo, an established and well-known service provider. - [SAFE]: Sensitive credentials like
KLAVIYO_PRIVATE_KEYare accessed via environment variables rather than being hardcoded, adhering to standard security practices. - [INDIRECT_PROMPT_INJECTION]: The skill defines workflows for processing external data and creating email templates, which constitutes a potential surface for indirect prompt injection findings.
- Ingestion points: The
createTemplateandcreateEventmethods inreferences/implementation.mdaccept HTML content and event properties that may contain untrusted data. - Boundary markers: No explicit instruction delimiters or 'ignore' directives are included in the code templates.
- Capability inventory: The skill has the capability to perform network operations and file system access via the enabled agent tools.
- Sanitization: The implementation relies on the Klaviyo API's own sanitization and rendering logic for template processing.
Audit Metadata