Skip to main content
Last updated on
Tags:
Options
Open in ChatGPTAsk ChatGPT about this pageOpen in ClaudeAsk Claude about this page

Run the n8n Demo

Spin up a fully configured demo on your machine — n8n with OpenBox governance pre-wired, a sample workflow, and a local LLM via Ollama.

Prerequisites​

  • Docker Desktop running
  • Ollama installed with at least one model pulled:
ollama pull llama3.2
ollama serve

Clone and Configure​

git clone https://github.com/OpenBox-AI/n8n-openbox-poc.git
cd n8n-openbox-poc

Copy .env.example to .env and set your API key:

.env
OPENBOX_API_KEY=obx_live_your_key_here

Start​

macOS / Linux:

./start.sh

Windows:

start.bat

This starts n8n and Postgres via Docker, installs the OpenBox node, creates a demo account, and activates the showcase workflow. Takes about a minute on first run.

Open n8n and Chat​

Navigate to http://localhost:5678 and log in:

  • Email: admin@openbox.ai
  • Password: Openbox123

Open the OpenBox SDK Showcase workflow and click Chat. Each message passes through full OpenBox governance before reaching the LLM.

See It in the Dashboard​

Open the OpenBox Dashboard:

  1. Navigate to Agents → click your agent
  2. On the Overview tab, find the session that corresponds to your workflow run
  3. Click Details to open the Event Log Timeline
  4. Scroll through the timeline — you'll see every governance stage:
    • Pre-call governance decision on your input
    • The LLM call with request and response
    • Post-call governance decision on the output
    • The final verdict returned to n8n

What Just Happened?​

When you chatted with the agent, the OpenBox govern() function:

  • Checked the input before the LLM call — your message was evaluated against configured guardrails and policies (PII detection, toxicity filtering, banned terms)
  • Ran your LLM call — the governed input was passed to the Ollama model
  • Checked the output after the LLM call — the LLM response was evaluated against post-call guardrails and policies
  • Recorded a governance decision for every stage — giving you a complete audit trail in the dashboard

Reset​

To tear down and restart from scratch:

docker compose down -v --rmi local
  • -v removes volumes (database data, etc.)
  • --rmi local removes images that were built locally, not pulled from a registry

Then run ./start.sh (macOS/Linux) or start.bat (Windows) again.

Next Steps​