n8n-mcp-tools-expert
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains high-priority steering instructions that direct the agent to always consult specific guides before calling n8n-mcp tools. These instructions are designed to prevent formatting errors and are aligned with the skill's intended purpose.
- [PROMPT_INJECTION]: Potential for Indirect Prompt Injection.
- Ingestion points: The agent ingests external workflow data and user-provided descriptions via tools such as
n8n_get_workflow,n8n_audit_instance, andn8n_generate_workflow. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when the agent is processing workflow JSON that could contain malicious comments or metadata.
- Capability inventory: The agent has extensive write and execution capabilities via
n8n_create_workflow,n8n_update_partial_workflow, andn8n_deploy_template, allowing it to modify code nodes and connections. - Sanitization: While the
n8n_audit_instancetool provides automated secret detection and masking, it does not explicitly handle potential injection patterns embedded within workflow data. - [DATA_EXFILTRATION]: The skill provides tools for managing credentials and auditing security. The documentation explicitly states that secret data is stripped from tool responses and masked in reports, which is a significant security feature that prevents accidental data exposure.
Audit Metadata