n8n-code-python

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No instructions to override agent behavior or bypass safety filters were found in the documentation.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or exfiltration patterns were detected. The skill correctly notes that external networking libraries like 'requests' are unavailable in the target environment.
  • [Indirect Prompt Injection] (LOW): The documentation describes handling external data from webhooks, which represents a standard ingestion surface.
  • Ingestion points: Untrusted data enters the context via _input.all() and _json['body'] as described in README.md.
  • Boundary markers: No specific delimiters or instruction-ignoring warnings are present in the code snippets.
  • Capability inventory: Execution is restricted to the n8n Python sandbox which lacks external networking and OS-level command capabilities.
  • Sanitization: The skill includes examples of defensive coding, such as using try/except for JSON parsing and .get() for dictionary access to prevent runtime errors.
  • [Remote Code Execution] (SAFE): The skill explicitly warns users that external packages cannot be installed (ModuleNotFoundError), reinforcing the platform's security constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:44 PM