n8n
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill documentation creates a surface for indirect prompt injection as it facilitates building workflows that ingest untrusted data while possessing high-privilege capabilities.\n
- Ingestion points: untrusted data enters the context via Webhook Trigger (n8n-nodes-base.webhook) and Form Trigger (n8n-nodes-base.formTrigger) nodes as documented in workflow-reference.md.\n
- Boundary markers: no specific instructions or delimiters are provided to isolate external data from internal logic or prompt expressions.\n
- Capability inventory: n8n workflows can execute shell commands via the Execute Command node, run arbitrary logic via Code nodes, and make network requests via HTTP Request nodes (SKILL.md, workflow-reference.md).\n
- Sanitization: the instructions do not mandate sanitization or validation of external content before interpolation into workflow parameters or code nodes.\n- [COMMAND_EXECUTION]: The skill describes the use of the n8n Execute Command node, which allows for the execution of arbitrary shell commands on the system hosting the n8n instance.\n- [REMOTE_CODE_EXECUTION]: The skill provides patterns for using n8n Code nodes which support arbitrary JavaScript and Python execution within the workflow pipeline.\n- [EXTERNAL_DOWNLOADS]: The skill references setup via the official n8n npm package, official Docker image (docker.n8n.io), and the n8n-nodes-starter repository on GitHub.
Audit Metadata