n8n Nodes

One node. Instant platform connection

Integration Trigger Node

Add this trigger node to your workflow and it automatically registers as an Employee on the AgentBrains platform, using the same name as your workflow.

The trigger node also pulls in the Employee's Campaign and Config fields automatically, so you can build agents that your clients can manage directly from the AgentBrains interface. Change a campaign brief or update a config setting on the platform, and your n8n workflow picks it up on the next run — no redeployment needed.

Deterministic retrieval. No guessing

Knowledge Base Node

Sometimes you don't want your agent to search and guess. You want it to read the exact file you created.

Use simple dropdown menus to select specific folders or files from your AgentBrains Knowledge Base. The node returns clean, parsed Markdown or JSON directly into your workflow — so the LLM receives exactly the data it needs to answer correctly.

Best for retrieving price lists, technical data sheets, instruction manuals, or strict company policies.

Infinite long-term memory. Zero setup.

RAG Node - Text Retrieval

Turn your entire Knowledge Base into a searchable brain. This node works as a standard AI Tool you can plug directly into your  AI Agent nodes in n8n.
No need to set up Pinecone, manage embeddings, or write Python scripts — just add the node as a tool. When a user asks a broad question like *"How do I troubleshoot the phone XXX?"*, the agent queries your entire Knowledge Base and retrieves the most relevant context.

Combine this with the Knowledge Base Node to give your agent both broad knowledge and pinpoint accuracy.

Show, don't just tell.

RAG Node - Image Retrieval

Standard AI agents are stuck in a text-only world. This node lets your n8n workflow fetch and display visual assets from your Knowledge Base.

Add it as a tool in your AI Agent's toolbelt. When a user asks to see something, the agent triggers the node and uses intelligent labels generated during ingestion (like *"Wiring Diagram for Model X"* or *"Red vs. Blue Color Comparison"*) to find the right image — no filename needed.

The node returns a direct, hosted URL that renders immediately in the chat, so your agent can share diagrams, product shots, or visual proof without leaving the conversation.

Validate your flows without rewriting them.

Synthetic QA Node

Testing agents in n8n usually means manually triggering workflows and reading JSON outputs. The QA Node changes that by connecting your workflow directly to the AgentBrains Synthetic User engine.

Drop this node into your existing flow — it doesn't alter your logic or change your output. It simply acts as a bridge to the AgentBrains testing environment. From there, spin up simulated personas like "Angry Customer" or "Curious Lead" that interact with your n8n workflow automatically.

View results as quantifiable data: aggregate quality scores (0–100) and pass/fail metrics for every test run, all visible in the AgentBrains dashboard.

Workflow samples