Descriptions:
Bart Slodyczka demonstrates how to connect OpenClaw — an open-source AI agent platform — to n8n for bidirectional communication, enabling complex background workflows to run without surfacing every message in the user-facing conversation thread. The integration uses Telegram as the front-end interface and n8n webhooks as the back-end processing layer, allowing the OpenClaw agent to dispatch events to n8n and receive structured responses that inform its next actions silently, as background context.
A central focus of the video is security architecture. Bart walks through a Docker-based deployment on Digital Ocean that routes all external traffic through Caddy — a reverse proxy acting as a perimeter security layer — before any request reaches n8n or OpenClaw. This design isolates the agent from direct internet exposure and positions n8n’s security nodes to screen incoming messages from channels like Gmail or WhatsApp for prompt injection before they reach the assistant. Bart explicitly notes the setup is suitable for testing but recommends further hardening before production use.
The deployment is deliberately simple: a $24/month Digital Ocean droplet running n8n, Caddy, and OpenClaw together in a single Docker container, configured via SSH in under 15 minutes. Bart also demonstrates a practical use case — triggering an n8n loop to call 20 job candidates via Retell AI phone calls, then reporting a structured summary back into the Telegram session. For builders looking to extend OpenClaw with n8n automation while adding infrastructure-level security, this video provides an end-to-end reference deployment.
📺 Source: Bart Slodyczka · Published February 02, 2026
🏷️ Format: Tutorial Demo
