• SUCCESS
  • ERROR
  • MANUAL
  • LOG

How to process loyalty reward redemptions with your AI

Connect Zinrelo to your agent to handle reward fulfillment. Once connected, your AI manages the actual redemption process for your customers.

Ask AI about this page

Short answer

How can my AI handle loyalty redemptions?

Your agent uses the redeem_reward tool to process specific requests for coupons or rewards. This allows you to move from a customer request to a completed redemption immediately. You'll see the results in your Zinrelo dashboard.

Redemption outcomes

Where redemptions land.

The results of your agent's actions.

SUCCESS

Reward issued

The tool completes the transaction. The customer gets their coupon or reward instantly.

ERROR

Fulfillment failed

If a reward is expired or points are low, your agent reports the error. You'll know exactly why it failed.

MANUAL

Human review needed

If a request hits a logic limit, the agent flags it for you. You handle these cases in Zinrelo.

LOG

Audit trail updated

Every action is recorded. You can verify the redemption history in your loyalty system.

The workflow

What your AI does when a redemption request arrives.

Your agent acts as the bridge between your chat and the Zinrelo backend.

  1. Identify the request

    Your agent parses the conversation to find the specific reward or coupon the customer wants.

    redeem_reward

  2. Validate eligibility

    The agent checks if the user has enough points for the requested item.

    redeem_reward

  3. Execute redemption

    The agent triggers the fulfillment in Zinrelo.

    redeem_reward

  4. Confirm status

    The agent tells the customer if the redemption worked or if there was a problem.

    redeem_reward

Try it

Copy these to start.

Use these prompts to trigger redemptions immediately.

Starting points

Swap the bracketed info for your specific customer details or reward names.

Zinrelo Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_AratjXjIrNe435yir0rGqwDlmZZdTjbFgj4FW3dU/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Zinrelo capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "zinrelo-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_AratjXjIrNe435yir0rGqwDlmZZdTjbFgj4FW3dU/mcp"
    }
  }
}
Copy into chat04

  • Redeem the 20% off summer coupon for customer account #8829.

  • Process a reward redemption for the 'Free Coffee' perk for user jdoe@email.com.

  • Check if I can redeem my loyalty points for the deluxe gift set right now.

  • Help this customer redeem their earned reward from the loyalty program.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Zinrelo once, then ask.

Link your Zinrelo account to Vinkius. Your credentials stay encrypted, and you can start managing rewards through any MCP-compatible client.

Connect Zinrelo to your AI

FAQ

How this task behaves

  • 01

    Can the AI create new loyalty programs?

    No. The agent can only process redemptions for existing rewards and coupons.

  • 02

    Does the AI change my customer's point balance?

    Yes, when the redeem_reward tool is used, points are deducted as part of the fulfillment.

  • 03

    Can the AI delete a customer from Zinrelo?

    No. There are no tools available to delete or modify customer profiles.

  • 04

    What happens if a coupon code is invalid?

    The agent will receive an error from the tool and inform you that the redemption failed.

  • 05

    Can I use this to send marketing emails?

    No. This connector is strictly for handling reward and coupon redemptions.

  • More questions about Zinrelo? See everything the Zinrelo Connector can do

Connect Zinrelo to Claude, Cursor, ChatGPT & more