palveron

n8n community nodes for Palveron AI Governance — verify prompts, manage agents, and enforce compliance policies in your n8n workflows.

Package Information

Released: 5 mo ago
Downloads: 1 weekly / 126 monthly
Latest Version: 1.0.0
Author: Palveron

Available Nodes

Documentation generating…

PALVERON
AI Governance — verify prompts, lookup traces, check agent status

Documentation

n8n-nodes-palveron

Official n8n community nodes for the Palveron AI Governance Gateway

npm version npm downloads License: MIT Documentation

Verify prompts, list active policies, and pull governance audit trails into any n8n workflow. Drop the Palveron node anywhere a prompt is built, branch on ALLOWED / BLOCKED / MODIFIED, and ship compliance-ready automation in minutes.

  • Community-node-package compliant — installable from the n8n UI without code
  • First-class credential type — API key + on-prem base URL
  • Per-call audit trail — every operation returns the Palveron trace ID

Installation

From the n8n UI

Settings → Community Nodes → Install → n8n-nodes-palveron.

Self-hosted / manual

cd ~/.n8n/nodes
npm install n8n-nodes-palveron

Then restart n8n.

Credentials

Create a Palveron API credential:

Field Value
API Key Your pv_live_… or pv_test_… key
Base URL https://gateway.palveron.com (or your self-hosted gateway)

The credential is reused by every Palveron node in a workflow.

Node operations

The Palveron node ships with three operations:

Operation Purpose
Verify Prompt Run a prompt through the gateway. Output exposes decision, output, reason, traceId, findings, and latencyMs. Branch on decision.
List Policies Fetch the project's active guardrails (id, name, prompt preview, environment). Useful for compliance dashboards.
Check Health Gateway readiness probe. Fail-fast workflows can guard on this before they call any model.

Example workflow

Webhook → Palveron (Verify Prompt) → IF decision == BLOCKED → Slack alert
                                  └→ OpenAI (Chat Completion) → Response

The OpenAI node always reads Palveron → output instead of the raw prompt, so anything the gateway redacted (PII, secrets) never reaches the model.

Requirements

  • n8n 1.0 or newer
  • Node.js 18 or newer
  • A Palveron account (free tier works for evaluation)

Links

License

MIT — Copyright © 2026 Palveron.

Discussion