delonet-n8n-architecture

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use n8n's Execute Command nodes to invoke host-side CLI utilities, such as bb-emit for event publishing and secure-source for file archiving. It also describes managing local service processes using pm2 restart n8n and setting environment variables like PM2_HOME to control the service manager.\n- [DYNAMIC_EXECUTION]: The documentation details a process for building, linking, and installing custom Node.js packages into the n8n application directory (~/.n8n/nodes). This mechanism allows for the runtime extension of the agent's capabilities via newly compiled code.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines an architecture for ingesting data from external sources, presenting a potential surface for indirect injection.\n
  • Ingestion points: Inbound Plane webhooks at https://n8n.delo.sh/webhook/plane and NATS message bus subscriptions.\n
  • Boundary markers: The skill requires strict raw-body HMAC signature verification using secrets managed through 1Password (op:// references) to ensure payload integrity.\n
  • Capability inventory: Configured workflows include shell command execution, file system operations (writing to a vault), and network communication.\n
  • Sanitization: Instructions focus on integrity validation and normalization according to predefined schemas, providing architectural protection against unauthorized data modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:55 AM
Security Audit — agent-trust-hub — delonet-n8n-architecture