Red-team AI agents with jailbreak probes using Groq, Discord, and data tables
Ready for 3.0Ready for 3.0. This template uses no nodes that n8n 3.0 removes. No removed nodes, versions or helpers found. Static check, last run 2026-10-19.
What this workflow does
This AI workflow has 43 nodes and runs manually, when you click “Execute workflow” (plus 2 other triggers). Along the way it connects Groq Chat Model, Data table and Discord.
How it runs
Trigger: Manually, when you click “Execute workflow” (plus 2 other triggers).
- Create Probe Table (Data table): create (table)
- Set Red Team Policy (Edit Fields (Set)): sets 3 fields: canary_token, max_tool_calls, escalate_below
- Set Digest Policy (Edit Fields (Set)): sets 2 fields: lookback_days, post_when_quiet
- Create Run Table (Data table): create (table)
- Load The Probes (Data table): get
- Read The Runs (Data table): get
- Create Result Table (Data table): create (table)
- Select The Probes (Code): runs custom JavaScript code
- Summarise The Trend (Code): runs custom JavaScript code
- Create Baseline Table (Data table): create (table)
…plus 25 more steps. See the full node list below.
Services it uses
Before you upgrade to n8n 3.0
Nothing in this workflow is removed in n8n 3.0, so it should import and run after the upgrade.
Nodes used (16)
- Manual Trigger
n8n-nodes-base.manualTrigger - Data table
n8n-nodes-base.dataTable - Code
n8n-nodes-base.code - n8n Form Trigger
n8n-nodes-base.formTrigger - Edit Fields (Set)
n8n-nodes-base.set - If
n8n-nodes-base.if - Loop Over Items (Split in Batches)
n8n-nodes-base.splitInBatches - AI Agent
@n8n/n8n-nodes-langchain.agent - Groq Chat Model
@n8n/n8n-nodes-langchain.lmChatGroq - Data Table Tool
n8n-nodes-base.dataTableTool - Code Tool
@n8n/n8n-nodes-langchain.toolCode - Basic LLM Chain
@n8n/n8n-nodes-langchain.chainLlm - Structured Output Parser
@n8n/n8n-nodes-langchain.outputParserStructured - n8n Form
n8n-nodes-base.form - Discord
n8n-nodes-base.discord - Schedule Trigger
n8n-nodes-base.scheduleTrigger
From the author
Quick overview This workflow red-teams an AI agent by replaying stored jailbreak and safety probes against a Groq-hosted model, scoring behavior from tool calls and responses, saving results to n8n Data Tables, and notifying a Discord channel plus a weekly…