ReferralCandy MCP. Process referrals, track rewards, and manage advocates in chat.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
ReferralCandy connects your AI agents directly to your affiliate and referral programs. It gives your agent full control over tracking advocates, registering purchases, managing campaigns, and analyzing performance metrics.
Use it to automatically process pending rewards, list top referrers by period, or send direct campaign invites without logging into the platform.
What your AI agents can do
Check referralcandy status
Verifies that the connection between your AI agent and ReferralCandy is active and working.
Get campaign
Retrieves full details for a specific referral marketing campaign by its ID or name.
Get referral
Pulls all specific data points about an individual referred user.
Retrieve overall program statistics and detailed lists of all referrals, advocates, and campaigns.
Register a purchase order directly to the system, automatically queuing rewards for advocates involved in the sale.
List all pending payouts and review current reward structures to ensure timely payments to advocates.
Identify the top-performing advocates, list all registered advocates, or retrieve a specific advocate's profile details.
List active campaigns, view overall program stats, and send targeted referral invitations to users.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 CompatibleWaiting for input…
ReferralCandy MCP Server: 16 Tools for Marketing Automation
Use these tools to track program metrics, register sales, list top advocates, and automate reward payouts via natural language prompts.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using ReferralCandy on Vinkius019dd14acheck referralcandy status
Verifies that the connection between your AI agent and ReferralCandy is active and working.
019dd14aget campaign
Retrieves full details for a specific referral marketing campaign by its ID or name.
019dd14aget referral
Pulls all specific data points about an individual referred user.
019dd14aget referrer
Fetches the profile and details for a specific advocate or referring user.
019dd14aget stats
Gathers overall, high-level performance statistics for your entire referral program.
019dd14aget top referrers
Identifies and lists the best advocates who have generated the most referrals or revenue recently.
019dd14alist campaigns
Provides a full list of every campaign currently set up in your referral program.
019dd14alist invites
Displays all past and current referral invitations that have been sent out by the system.
019dd14alist pending rewards
Checks for payouts or rewards that are due but have not yet been processed in the system.
019dd14alist purchases
Lists all purchases made by referred users, including order IDs and attribution details.
019dd14alist referrals by period
Filters and lists referrals, restricting results to a specific date range or period you specify.
019dd14alist referrals
Generates a comprehensive list of every referral recorded in the program's history.
019dd14alist referrers
Gets a complete directory list of all advocates who have ever signed up into the program.
019dd14alist rewards
Displays a comprehensive catalog and status list of all reward types available in your system.
019dd14aregister purchase
Manually logs a purchase event, which triggers the automatic calculation and queuing of rewards for the advocate.
019dd14asend invite
Sends a new referral invitation directly to a specified user or group within your program.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with ReferralCandy, then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by ReferralCandy. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 16 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Manually tracking referrals and rewards across multiple tabs is a nightmare.
Today, running end-of-month reports means opening your platform dashboard, filtering by date range. Then you switch to the 'Advocates' tab to find top performers, copy their IDs into a spreadsheet, and then finally jump over to the 'Rewards' section just to see if the payouts were correctly queued—it takes at least five clicks and three tabs.
With this MCP server, you ask your agent one thing: 'What rewards are pending for November?' The agent runs `list_pending_rewards` and gives you a clean summary. It handles the navigation and data correlation; you just get the answer.
ReferralCandy MCP Server lets you manage rewards via API.
Before, if an advocate made a purchase, the reward payout was often delayed or required manual follow-up. You'd have to check the order data against the reward configuration and then manually confirm the pending status in the backend. This process is slow and prone to human error.
Now, you tell your agent: 'Process this $200 sale for order #123.' The agent executes `register_purchase`. It handles the attribution and queues the payout instantly. That's how fast and reliable it's now.
What you can do with this MCP connector
ReferralCandy gives your AI agent full control over managing your entire affiliate and referral program. You'll use this server to let your agent handle everything from tracking a new sign-up to processing payments, all without you having to log into the platform yourself.
Analyzing Program Performance
You can get a complete picture of how your referrals are doing across several tools. Your agent pulls overall performance statistics using get_stats. To see who's driving sales, it identifies and lists the top-performing advocates via get_top_referrers, or you can pull the full directory list of every advocate signed up with list_referrers.
When you need deep data on a single user, your agent pulls all specific details about an individual referred customer using get_referral and fetches the profile information for a specific referring advocate through get_referrer.
To track sales activity, your agent reads every referral recorded in the program's history with list_referrals, or it filters results to focus on a narrow time window by calling list_referrals_by_period. For purchase data, you can list all purchases made by referred users—including order IDs and attribution details—using list_purchases. You'll also find that your agent checks the activity of your system connection using check_referralcandy_status to make sure everything's running smoothly.
Managing Transactions and Rewards
When a sale happens, you don't want to wait for manual processing. Your agent handles it instantly: calling register_purchase manually logs the purchase event. This action triggers the automatic calculation of rewards and queues up any necessary payouts for the advocate involved in that sale.
For payout management, your agent checks what needs paying out by running list_pending_rewards. You can also review your reward catalog and status list using list_rewards to ensure you're always set up with current incentives. If you need targeted marketing, your agent sends a new referral invitation directly to a specified user or group through send_invite, keeping your program active.
Controlling Campaigns and System Data
Your agent manages the underlying structure of your programs too. It provides full details for a specific referral marketing campaign using get_campaign, and it also fetches all existing campaigns when you call list_campaigns. If you need to know what invitations have been sent out, past or current, calling list_invites shows that history.
To keep track of cash flow, your agent can list every reward type available in the system using list_rewards, and it also gives a full inventory of all rewards that are due but haven't cleared yet by listing pending payouts via list_pending_rewards. You don't need to log in; your AI client does the heavy lifting for you.
019dd14a-c1c5-73a4-a889-bf59ce89aa8c How ReferralCandy MCP Works
- 1 Subscribe to the ReferralCandy server and paste your API keys (you find these in your developer settings).
- 2 Your AI agent uses a tool like
check_referralcandy_statusto verify connectivity and confirm credentials. - 3 The agent then executes multi-step commands, such as listing referrals by period (
list_referrals_by_period) followed by registering purchases (register_purchase).
The bottom line is that your AI client uses the API keys to execute complex marketing and reporting actions against your active ReferralCandy account.
Who Is ReferralCandy MCP For?
Growth hackers, Marketing Operations Managers, and E-commerce Analysts. Use this if you're tired of jumping between a dashboard, a spreadsheet, and the platform UI just to get a single metric or process one payout batch. This lets your agent handle the entire workflow conversationally.
Uses list_referrals and get_stats to pull weekly performance data, then uses list_pending_rewards to confirm which payouts need manual attention.
Runs list_purchases and correlates the resulting order IDs with campaign details using get_campaign to calculate effective ROI.
Uses send_invite or list_invites to quickly scale outreach, sending targeted campaigns to specific groups of advocates identified via list_referrers.
What Changes When You Connect
- Automate Reward Processing: Use
register_purchaseto log a sale. Your agent instantly queues the corresponding reward payout without you touching the UI. This moves transactions from manual data entry to conversational commands. - Deep Data Filtering: Don't just look at totals. Tools like
list_referrals_by_periodlet your agent filter referrals by exact dates, giving you precise metrics for quarterly reports that standard dashboards can't easily isolate. - Instant Advocate Intelligence: Forget scrolling through hundreds of users. The
get_top_referrerstool surfaces the highest-impact advocates immediately, allowing you to pivot directly into sending a retention bonus viasend_invite. - Program Health Check: Use
list_pending_rewardsandlist_rewards. Your agent reviews what's owed versus what's configured, flagging payout discrepancies before they become cash flow issues. - Campaign Lifecycle Management: Need to adjust outreach? The combination of
list_campaigns,get_campaign, andsend_invitemeans your agent can monitor, modify, and execute campaign rollouts—all in one chat session.
Real-World Use Cases
The End-of-Month Payout Audit
A marketing analyst needs to confirm all payouts for November. Instead of running five different reports, they ask their agent: 'List all pending rewards.' The agent runs list_pending_rewards and flags $50 in missing attribution. They then use get_referral on the involved user ID to find out why the referral wasn't fully tracked.
Launching a Targeted Advocate Campaign
The e-commerce manager wants to boost sales from top users. They ask their agent: 'Who are my best advocates this quarter?' The agent runs get_top_referrers. Based on the list, the manager asks the agent to send a special incentive invite using send_invite.
Manually Logging a Bulk Sale
A sales representative processes an offline corporate sale. Instead of waiting for web data sync, they tell their agent: 'Log $500 for order #XYZ.' The agent executes register_purchase, ensuring the reward is immediately queued and attributed to the correct advocate.
Reviewing Historical Program Performance
The head of growth needs metrics for Q2. They ask their agent: 'Show me all referrals from April 1st to June 30th.' The agent uses list_referrals_by_period, providing a clean, filtered dataset without needing manual date range adjustments.
The Tradeoffs
Trying to get all data in one shot
Asking the agent: 'Give me everything about referrals and purchases.' The agent can't guess your focus, resulting in a massive, unusable dump of raw data.
→
Be specific. Start by asking for an overview using get_stats. Then narrow it down. If you want to check rewards, use list_pending_rewards; if you need sales records, use list_purchases.
Confusing advocates and referrals
Asking the agent: 'List all people.' This is too broad. You'll get an unmanageable list of every user, including non-advocates.
→
Always use dedicated tools. If you want a directory of users who can refer, run list_referrers. If you need to see historical activity, run list_referrals.
Assuming automatic attribution
Thinking that just logging an order number is enough. The system needs the explicit connection point.
→
If you are manually entering a sale, always use register_purchase. This tool forces the agent to attribute the sale and queue the reward correctly.
When It Fits, When It Doesn't
Use this server if your primary pain is workflow orchestration within your referral program. You need an agent that can execute multi-step processes: 'List top referrers' -> then 'Send invite to them' -> and finally 'Check pending rewards.'
Don't use it if you just need raw data exports. If all you want is a massive CSV of every single referral, downloading the list directly from your platform might be faster. However, if that raw data needs to trigger an action (like updating a record or sending an invitation), this server is required.
If your goal is complex financial modeling (e.g., correlating referral revenue with external ad spend), you'll need tools outside of marketing automation—you'd use the output from get_stats as input into another data pipeline.
Common Questions About ReferralCandy MCP
How do I find out which advocates are doing well with get_top_referrers? +
The get_top_referrers tool lists your highest-performing advocates. It ranks them based on key metrics, letting you know exactly who to contact for a bonus or recognition.
Can I list referrals from only last month using list_referrals_by_period? +
Yes. list_referrals_by_period lets you filter the entire referral history by a specific start and end date, making it perfect for monthly reporting.
How do I make sure rewards are paid out correctly? list_pending_rewards helps? +
The list_pending_rewards tool shows exactly what payouts are ready to go. This lets you confirm the system is keeping up with sales and that no payments are stuck.
What if I need to manually log a purchase? Should I use register_purchase? +
Yes, register_purchase is the tool for it. It ensures that even if the purchase wasn't made through the standard checkout flow, the corresponding reward payout gets queued.
Before I run any queries, how can I use check_referralcandy_status to confirm my API keys work? +
Running check_referralcandy_status confirms your connection and credentials immediately. It returns a simple success code if the server recognizes your keys and has live access. If you get an error, it usually means the key is expired or wasn't correctly entered in Vinkius.
If I run list_purchases for thousands of records, will I get all of them? Does it handle pagination? +
The system handles large data sets by checking for paginated results. If the initial call doesn't return everything you expect, your AI client should automatically request the next page using the provided cursor or offset token to gather the complete list.
What happens if I use get_referral with a referral ID that doesn't exist? +
The tool returns a specific 'Not Found' error code and an explanatory message. This tells your agent immediately that the provided ID is invalid, preventing it from running unnecessary follow-up logic.
I need to send out many invites quickly; are there any limits when I use send_invite? +
Yes, API rate limits apply. If you try sending a massive batch of invitations in rapid succession, the system will return a 429 'Too Many Requests' error. You’ll need to implement a short delay or queue the requests.
Can my AI automatically register a new purchase to trigger a referral reward? +
Yes! Use the register_purchase tool providing the buyer's email, order amount, and order ID. Your agent will send this data to ReferralCandy, triggering the referral attribution and payout flow instantly.
How do I easily find out who my best performing advocates are? +
Simply ask the agent to run the get_top_referrers action. It will compile your highest-performing ambassadors ranked by total successful referrals and the total revenue they generated for your store.
Can I check if a specific customer has any pending referral rewards? +
Absolutely. By using the list_pending_rewards tool and filtering by the customer's email, your agent can instantly fetch their reward status and payout timeline without you needing to open the dashboard.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.