n8n-code-javascript
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONDYNAMIC_EXECUTIONPERSISTENCE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents how to process untrusted data from webhooks and external files, creating a surface for injection attacks. Ingestion points: Processing webhook data via $json.body and markdown content (DATA_ACCESS.md, COMMON_PATTERNS.md). Boundary markers: Includes reporting patterns that use headers and dividers to delimit content (COMMON_PATTERNS.md Pattern 10). Capability inventory: Capabilities include network operations via this.helpers.httpRequest() and dynamic module loading via require() (BUILTIN_FUNCTIONS.md). Sanitization: Demonstrates data normalization for CRM inputs and regex-based extraction to filter specific patterns (COMMON_PATTERNS.md Pattern 5).
- [DATA_EXFILTRATION]: Describes the use of this.helpers.httpRequest() to send data to external URLs. The skill includes appropriate guidance on managing authentication safely by using the platform's credential system instead of hardcoding secrets or relying on environment variables.
- [DYNAMIC_EXECUTION]: Documents the use of require() for loading Node.js modules like crypto and the Buffer global for base64 operations. These capabilities are gated by environment configuration but represent potential attack surfaces if misused.
- [PERSISTENCE]: Explains the use of $getWorkflowStaticData() to maintain state across different workflow executions. This is a standard n8n platform feature for state management but allows for data persistence across sessions.
Audit Metadata