using-n8n-mcp-skills

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process workflow definitions and node configurations from external n8n instances using tools like n8n_get_workflow and get_node. This data represents an attack surface where instructions embedded in workflow names, node descriptions, or code blocks could potentially influence agent behavior. Do not include severity levels in the findings. Use markdown formatting with bold category headers and nested bullet points for evidence details.
  • Ingestion points: Data is received through n8n_get_workflow, get_node, search_nodes, and tools_documentation as described in SKILL.md.
  • Boundary markers: The instructions lack explicit delimitation or framing for the ingested workflow data to prevent the agent from interpreting embedded text as instructions.
  • Capability inventory: The skill utilizes powerful tools across all scripts, including n8n_create_workflow, n8n_update_partial_workflow, n8n_manage_credentials, and n8n_test_workflow (which executes real nodes).
  • Sanitization: There are no specific instructions provided to the agent for sanitizing or escaping content retrieved from the n8n instance before processing it.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 05:43 PM
Security Audit — agent-trust-hub — using-n8n-mcp-skills