{"product_id":"red-team-ai-agents-with-jailbreak-probes-groq-discord-n8n","title":"Red-Team AI Agents with Jailbreak Probes (Groq, Discord, n8n)","description":"\u003ch3\u003eRed-team your Groq AI agent safely—by replaying jailbreak probes and scoring behavior end-to-end\u003c\/h3\u003e\n\u003cp\u003eThis n8n workflow automates red-teaming of an AI agent by replaying stored jailbreak and safety probes against a Groq-hosted model, scoring tool use and responses for canary leakage and policy compliance, then saving results to n8n Data Tables and notifying a Discord channel (plus a weekly digest).\u003c\/p\u003e\n\n\u003ch3\u003eWhat this workflow does\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003eOne-time manual setup:\u003c\/strong\u003e creates the required n8n \u003cstrong\u003eData Tables\u003c\/strong\u003e, loads a default probe set, and inserts demo customer records.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eStart a red-team run from an n8n Form:\u003c\/strong\u003e select a probe suite, choose the maximum number of probes, and set the target system prompt. A `canary` build key is appended automatically to detect leakage.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eLoad and queue probes for regression:\u003c\/strong\u003e enables probes (optionally filtered by suite), reads a stored baseline, and attaches each probe’s prior verdict so runs can be compared over time.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eExecute probes with tool constraints:\u003c\/strong\u003e sends each probe’s attack prompt to an \u003cstrong\u003en8n AI Agent backed by Groq\u003c\/strong\u003e, allowing only a customer-record lookup tool and a simulated external messaging tool while capturing intermediate tool steps.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eDeterministic scoring and review:\u003c\/strong\u003e checks for canary leakage, forbidden tool use, compliance markers, record overreach, refusal behavior, tool-call caps, and blocked\/errored outputs. A Groq reviewer can downgrade suspicious “PASS” answers to “REVIEW” using verbatim quotes.\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003ePersist and notify:\u003c\/strong\u003e writes the run report, per-probe results, and updated baseline to Data Tables, posts failures to \u003cstrong\u003eDiscord\u003c\/strong\u003e, and displays a completion page (including a weekly Discord digest).\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eUse cases\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003eContinuously validate that an automation engineer’s Groq-powered agent cannot leak a canary key or access data beyond allowed tools.\u003c\/li\u003e\n  \u003cli\u003eSaaS operators run scheduled regression checks against jailbreak and safety probes after prompt\/tool changes.\u003c\/li\u003e\n  \u003cli\u003eDebug tool-call caps and failure modes by capturing blocked\/errored outputs per probe.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eTechnical details\u003c\/h3\u003e\n\u003cul\u003e\n  \u003cli\u003e\n\u003cstrong\u003en8n nodes:\u003c\/strong\u003e \u003cem\u003eif\u003c\/em\u003e, \u003cem\u003eset\u003c\/em\u003e, \u003cem\u003ecode\u003c\/em\u003e, \u003cem\u003eform\u003c\/em\u003e, \u003cem\u003ediscord\u003c\/em\u003e, \u003cem\u003edata table\u003c\/em\u003e\n\u003c\/li\u003e\n  \u003cli\u003e\n\u003cstrong\u003eIntegrations:\u003c\/strong\u003e \u003cstrong\u003eGroq\u003c\/strong\u003e for the AI agent and Groq-based reviewer; \u003cstrong\u003eDiscord\u003c\/strong\u003e for alerts and weekly digest; \u003cstrong\u003en8n Data Tables\u003c\/strong\u003e for baseline and run persistence.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eKeywords:\u003c\/strong\u003e n8n automation workflow, red-team AI agents, Groq, Discord alerts, jailbreak probes, AI safety regression testing, Data Tables.\u003c\/p\u003e","brand":"N8N Commerce","offers":[{"title":"Default Title","offer_id":45967551725747,"sku":"N8N-19264","price":27.99,"currency_code":"GBP","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0749\/6279\/6723\/files\/oLKEbt3_7_ufCLjlDRPPD_SqVpe5nc.png?v=1788685723","url":"https:\/\/buyflowscripts.com\/products\/red-team-ai-agents-with-jailbreak-probes-groq-discord-n8n","provider":"N8N Commerce","version":"1.0","type":"link"}