Skip to main content

​
Overview

This guide walks you through setting up Velatir to monitor your AI systems.

​
Step 1: Create a Project

  1. Log in to your Velatir dashboard
  2. Click Create Project
  3. Give your project a name and description
  4. Click Create

​
Step 2: Enable Policies

  1. Go to Policies in the dashboard
  2. Browse the available pre-built policies
  3. Enable the policies relevant to your use case
  4. Optionally create custom policies for your specific requirements

​
Step 3: Create a Workflow

  1. Go to Workflows in the dashboard
  2. Click Create Workflow
  3. Add nodes to your workflow:
    • Policy Trigger - Triggers when a policy matches
    • Human Intervention - Creates a review task for approval
    • Notification - Sends alerts to your team
    • Rejection - Rejects the trace
  4. Connect the nodes to define your approval flow
  5. Save and activate your workflow

​
Step 4: Connect a Channel

  1. Go to Channels in the dashboard
  2. Connect your preferred channel:
    • Slack - Approve directly in Slack
    • Microsoft Teams - Use adaptive cards
    • Email - Reply-to-approve workflow
  3. Human intervention nodes will now send approval requests to your connected channels

​
Step 5: Integrate Your AI System

Choose an integration method that fits your stack:

Browser Extension

Monitor LLM interactions across 200+ AI platforms

n8n Node

Add approval gates to n8n workflows

LangChain

Native middleware for LangChain agents

MCP Server

Model Context Protocol for AI agents

TypeScript SDK

Official TypeScript/JavaScript library

Python SDK

Official Python library

​
What’s Next?

Understand Traces

Learn about trace directions, sessions, and the data model

Configure Workflows

Build complex workflows with multiple triggers and conditions

Explore Policies

Understand pre-built and custom policy creation

Set Up Channels

Connect Slack, Teams, or email for approvals