n8n-code-javascript

Warn

Audited by Gen Agent Trust Hub on Feb 14, 2026

Risk Level: MEDIUMPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (MEDIUM): The skill instructions facilitate an Indirect Prompt Injection surface by detailing how to ingest untrusted data and perform network operations without accompanying security constraints.
  • Ingestion points: Webhook body access ($json.body) and HTTP request results are the primary entry points for untrusted data into the generated scripts.
  • Boundary markers: The skill does not provide patterns for using delimiters or boundary markers to isolate untrusted data within generated code logic.
  • Capability inventory: The skill describes capabilities including outbound network requests via $helpers.httpRequest and state modification through $getWorkflowStaticData.
  • Sanitization: While the skill mentions null checks for error prevention, it lacks guidance on sanitizing or escaping external inputs used in sensitive downstream operations.
  • [NO_CODE] (INFO): The analyzed file is documentation (README.md) and contains no executable code or scripts, functioning solely as instructional metadata for an AI agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 14, 2026, 12:42 AM