Auto-review n8n workflow JSON for risks and best practices with Groq

Webhook receives exported n8n workflow JSON → code node static validation → Groq AI risk assessment → returns structured JSON report

Who

Automation engineers or platform admins who need to batch-review n8n workflow quality and security risks

You get

Generate a structured assessment report with metrics, risk items, business impact, and fix recommendations for any n8n workflow JSON

You need

n8n

Steps

1. Configure Groq credentials and webhook trigger

   Open the template workflow in n8n. Select the Groq Analysis Chat Model node and add a groqApi credential with your API key from console.groq.com. Then select the When Workflow Received (Webhook) node, confirm HTTP Method is POST and Response Mode is responseNode; the default path is workflow—update the caller config if you change it.

2. Import and activate the workflow

   Import the full workflow.json into your n8n instance (or install directly from the n8n template library). Verify node connections: Webhook → Validate Workflow Data → If Validation Succeeded → true branch to Risk Assessment Agent, false branch to Generate Error Response → both routes finally merge into Send Analysis Response. Click Activate when intact.

   Setup guidance · Editorial suggestion。Setup assistant starter prompt; used to verify import completeness and credential configuration

   Copy prompt

   Please set up the n8n workflow review service using the provided workflow JSON: after import, verify the webhook path, validation code, AI Agent prompt, and structured output parser schema are intact. Test Groq credential connectivity before activating.

3. Send workflow JSON to the webhook

   Copy the activated webhook URL from n8n. In your caller (e.g., curl, Postman, or another workflow), construct a POST request with a JSON body containing a top-level workflow field whose value is the exported n8n workflow JSON. Example: {"workflow": {"name": "...", "nodes": [...], "connections": {...}}}.

4. Review the AI risk assessment result

   After successful validation, the Risk Assessment Agent invokes Groq (llama-3.3-70b-versatile) to analyze metrics and findings, formats through Parse Structured Output, and returns JSON. The response includes score, security, performance, cost, complexity, suggestions. If validation fails, it returns {"success": false, "error": "..."}.

   Execution prompt · Source text。Risk Assessment Agent node text parameter (n8n expression, starts with =)

   Copy prompt

   =You are a Senior n8n Automation Engineer.
   
   The workflow has already been analyzed.
   
   DO NOT invent issues.
   
   Use the findings below.
   
   Explain
   
   • Why it matters
   
   • Risk
   
   • Business impact
   
   • Best practice
   
   Return ONLY JSON.
   
   Metrics
   
   {{ JSON.stringify($json.metrics,null,2) }}
   
   Findings
   
   {{ JSON.stringify($json.findings,null,2) }}

Please set up the n8n workflow review service using the provided workflow JSON: after import, verify the webhook path, validation code, AI Agent prompt, and structured output parser schema are intact. Test Groq credential connectivity before activating.