Connect Mollie to Claude: Automate Subscriptions and Payouts

If you need to connect Mollie to Claude to automate subscription management, streamline payment links, or handle complex refund workflows, you need a Model Context Protocol (MCP) server. This server acts as the translation layer between Claude's natural language tool calls and Mollie's REST API. You can either spend weeks building and maintaining this infrastructure yourself, or use a managed integration platform like Truto to dynamically generate a secure, authenticated MCP server URL. If your team uses ChatGPT, check out our guide on connecting Mollie to ChatGPT or explore our broader architectural overview on connecting Mollie to AI Agents.

Giving a Large Language Model (LLM) read and write access to a sprawling financial ecosystem like Mollie is a significant engineering challenge. You have to handle OAuth 2.0 token lifecycles, map massive JSON schemas for payments and profiles to MCP tool definitions, and deal with Mollie's specific API quirks. Every time Mollie updates an endpoint or deprecates a field, you have to update your server code, redeploy, and test the integration. This guide breaks down exactly how to use Truto to generate a secure, managed MCP server for Mollie, connect it natively to Claude, and execute complex financial workflows using natural language.

The Engineering Reality of the Mollie API

A custom MCP server is a self-hosted integration layer. While the open MCP standard provides a predictable way for models to discover tools, the reality of implementing it against a payment gateway's API is painful. You are not just integrating a simple database - you are integrating highly sensitive financial rails.

If you decide to build a custom MCP server for Mollie, you own the entire API lifecycle. Here are the specific challenges you will face:

Idempotency and Rate Limit Transparency Financial APIs demand exactness. Mollie enforces strict rate limiting based on the endpoint category (e.g., creating payments versus listing profiles). When integrating manually, an HTTP 429 Too Many Requests response can break your AI agent if it isn't expecting it. Truto normalizes Mollie's rate limit headers into IETF standard formats (ratelimit-limit, ratelimit-remaining, ratelimit-reset). Truto does not absorb or retry these errors automatically - it passes the 429 directly to Claude. This is a massive advantage because it allows the model to see the ratelimit-reset window and autonomously apply its own backoff strategy rather than failing silently behind a proxy timeout.

Complex Pagination and Cursor Management Mollie relies heavily on HAL (Hypertext Application Language) _links for pagination. Passing raw _links.next.href strings to an LLM is a disaster waiting to happen - the model will often hallucinate URL parameters or attempt to construct its own pagination tokens. Truto abstracts this away, presenting a normalized limit and next_cursor schema to Claude. The tool description explicitly instructs the LLM to pass cursor values back unchanged, completely eliminating pagination hallucinations.

Deeply Nested Financial Objects A single Mollie payment object contains deeply nested objects for routing, captures, chargebacks, and localized payment method details. Building JSON-RPC tool schemas that accurately represent these nested payloads so an LLM can understand them requires writing thousands of lines of boilerplate validation code. Truto dynamically maps Mollie's resource schemas directly into MCP-compatible formats on the fly.

How to Generate a Mollie MCP Server with Truto

Truto dynamic tool generation works by reading the underlying integration documentation and automatically building JSON-RPC 2.0 endpoints for every available Mollie API operation.

You can create a Mollie MCP server using either the Truto UI or the API.

Method 1: Via the Truto UI

For teams that want a fast, no-code setup, the Truto interface provides a one-click deployment.

1. Log in to your Truto dashboard and navigate to the integrated account page for your Mollie connection.
2. Click the MCP Servers tab.
3. Click Create MCP Server.
4. Select your desired configuration (name, allowed methods, tags, and expiry).
5. Copy the generated MCP server URL.

Method 2: Via the Truto API

For platform engineers who need to programmatically provision AI access to Mollie for different tenants or internal microservices, you can generate MCP servers via the Truto API.

Make a POST request to the /integrated-account/:id/mcp endpoint:

curl -X POST https://api.truto.one/integrated-account/<your_integrated_account_id>/mcp \
  -H "Authorization: Bearer <your_truto_api_token>" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Mollie Subscription Agent",
    "config": {
      "methods": ["read", "write"],
      "tags": ["payments", "subscriptions"]
    }
  }'

The API will return a secure, hashed token URL:

{
  "id": "mcp_abc123",
  "name": "Mollie Subscription Agent",
  "url": "https://api.truto.one/mcp/a1b2c3d4e5f6...",
  "config": {
    "methods": ["read", "write"],
    "tags": ["payments", "subscriptions"]
  }
}

This URL contains a cryptographic token that securely maps to the specific Mollie tenant. It is fully self-contained.

How to Connect the MCP Server to Claude

Once you have your Truto MCP server URL, you need to register it with your Claude client. You can do this via the Claude UI for individual users, or via a configuration file for system-wide deployments.

Method A: Via the Claude UI

If you are using the Claude desktop app or web interface with Custom Connector support:

1. Open Claude and navigate to Settings.
2. Click on Integrations (or Connectors) and select Add MCP Server.
3. Paste the Truto MCP Server URL you generated in the previous step.
4. Click Add.

Claude will immediately perform a handshake with the Truto MCP server, requesting the tools/list endpoint to discover all available Mollie operations.

Method B: Via Manual Config File

For developers managing Claude Desktop locally, you can modify the claude_desktop_config.json file to pipe requests through the official MCP SSE (Server-Sent Events) transport. This is the preferred method for robust local development.

Update your configuration file (located at ~/Library/Application Support/Claude/claude_desktop_config.json on macOS or %APPDATA%\Claude\claude_desktop_config.json on Windows):

{
  "mcpServers": {
    "mollie_production": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sse",
        "https://api.truto.one/mcp/<your_secure_token>"
      ]
    }
  }
}

Restart Claude Desktop. The model now has real-time, authenticated read/write access to your Mollie instance.

Hero Tools for Mollie Operations

Truto automatically generates dozens of tools for Mollie. Below are 6 high-leverage "hero tools" that transform Claude from a simple chatbot into a capable financial operations agent.

create_a_mollie_payment_link

This tool allows the agent to instantly generate shareable checkout links. It is incredibly useful for support agents dealing with invoice disputes or sales teams finalizing custom quotes over chat.

Usage note: You must provide a description and a valid amount format (value and currency). The tool returns a paymentLink URL that you can immediately present to the user.

"The customer wants to upgrade their tier. Create a Mollie payment link for 50.00 EUR with the description 'Pro Tier Upgrade - December'. Give me the checkout URL."

get_single_mollie_payment_by_id

Retrieves the full lifecycle data of a specific transaction. Essential for investigating "where is my payment?" inquiries or checking if a transaction has settled.

Usage note: The id parameter requires the exact Mollie payment ID (e.g., tr_7UhSN1zuRU). It returns nested data including method, status, amount, and related routing.

"Check the status of payment tr_7UhSN1zuRU. Has it been captured yet, or is it still pending?"

create_a_mollie_payment_refund

Empowers Claude to autonomously execute refunds directly from the chat interface.

Usage note: Requires the payment_id. If you do not specify an amount, the tool will attempt a full refund. This tool should typically be placed behind a human-in-the-loop approval workflow for safety.

"The user was double-charged for order 99281. Locate the payment ID, verify the amount, and initiate a full refund."

create_a_mollie_customer_subscription

Allows the AI agent to attach recurring billing schedules to an existing Mollie customer. Perfect for self-serve onboarding flows where the AI acts as a concierge.

Usage note: Requires the customer_id, the amount object, the interval (e.g., "1 month"), and a description. The customer must already have a valid mandate on file.

"Set up a new subscription for customer cst_8wmqcHMN4U. The amount is 15.00 EUR per month, starting today, for the 'Basic Plan'."

list_all_mollie_settlements

Exposes the payout history from Mollie to your business bank accounts. Crucial for financial operations and automated reconciliation workflows.

Usage note: You can filter by balanceId or query year and month parameters. The tool handles cursor-based pagination automatically.

"List the last 5 settlements we received from Mollie. Give me the dates, the total amounts, and the status of each."

list_all_mollie_payment_chargebacks

Surfaces critical risk data. Allows the AI agent to proactively flag high-risk accounts or automatically compile data required to dispute a chargeback.

Usage note: Pass a specific payment_id to see if it was reversed, or list chargebacks generally to audit portfolio risk.

"Look up payment tr_Q2z1bz. Has a chargeback been initiated for this transaction? If so, what is the stated reason code?"

For a complete list of available operations, schemas, and required parameters, review the Mollie integration page.

Workflows in Action

Connecting Mollie to Claude unlocks the ability to chain these tools together into autonomous, multi-step workflows.

Use Case 1: The Autonomous Support Refund

Customer support teams waste countless hours toggling between Zendesk and the Mollie dashboard just to process simple refunds. With an MCP-enabled Claude, this entire flow happens via natural language.

"Customer cst_Jk92lX emailed us to say they accidentally purchased the annual plan instead of the monthly plan. Please find their most recent payment, verify it's the annual amount, and process a full refund."

1. Claude calls list_all_mollie_customer_payments passing cst_Jk92lX to retrieve the user's transaction history.
2. Claude identifies the most recent payment ID (tr_V9z2Lx) and checks the amount field to confirm it matches the annual plan pricing.
3. Claude calls create_a_mollie_payment_refund using tr_V9z2Lx, executing the actual return of funds.

The Outcome: The support agent receives immediate confirmation in the chat interface that the refund was successful, along with the refund ID, allowing them to reply to the customer in seconds.

sequenceDiagram
    participant Agent as Support Rep
    participant Claude as Claude Desktop
    participant Truto as Truto MCP
    participant Mollie as Mollie API

    Agent->>Claude: "Refund recent payment for cst_Jk92lX"
    Claude->>Truto: call list_all_mollie_customer_payments
    Truto->>Mollie: GET /v2/customers/cst_Jk92lX/payments
    Mollie-->>Truto: return payment array
    Truto-->>Claude: JSON response
    Claude->>Truto: call create_a_mollie_payment_refund
    Truto->>Mollie: POST /v2/payments/tr_V9z2Lx/refunds
    Mollie-->>Truto: return refund object
    Truto-->>Claude: confirmation
    Claude-->>Agent: "Refund processed successfully."

Use Case 2: Automated Subscription Recovery

When a customer asks to update their billing via chat, the agent usually has to generate a new link manually, send it over, wait for the mandate to clear, and then restart the subscription. Claude can orchestrate this entire lifecycle.

"User ID cst_P82nA wants to restart their failed subscription for the 'Enterprise Plan' at 500 EUR/month. Send them a checkout link to capture a new mandate, then tell me when it's done so we can initiate the subscription."

1. Claude calls create_a_mollie_payment_link for 500 EUR with the description 'Enterprise Plan Mandate'.
2. Claude returns the URL to the chat so it can be sent to the customer.
3. (Once the customer pays), the human types: "They paid it. Start the subscription."
4. Claude calls create_a_mollie_customer_subscription passing the customer_id and the monthly interval data.

The Outcome: An operation that normally requires cross-referencing documentation, waiting on webhooks, and manual data entry in the Mollie UI is reduced to two natural language prompts.

Security and Access Control

Exposing financial infrastructure to an LLM requires strict boundary setting. Truto provides four distinct layers of security at the MCP token level:

• Method Filtering (config.methods): You can restrict an MCP server to strictly read-only operations by passing ["read"] during server creation. This allows Claude to query balances and payments but physically prevents it from initiating refunds or transfers.
• Tag Filtering (config.tags): Group tools by business domain. For example, passing ["support"] might only expose ticketing and refund endpoints, hiding sensitive endpoints like organizational balances or API token generation.
• Additional Authentication (require_api_token_auth): By default, possessing the MCP URL is enough to access the tools. Setting this flag to true requires the connecting client to also pass a valid Truto API token in the headers, adding a secondary identity check.
• Automatic Expiration (expires_at): You can generate time-boxed servers by passing an ISO datetime. Once the timestamp is reached, Truto automatically triggers a cleanup alarm, purging the token from the KV store and database instantly.

Orchestrating Financial Operations with AI

Connecting Mollie to Claude using a managed MCP server removes the friction of building financial integration infrastructure. Instead of parsing rate limits, wrangling deeply nested JSON-HAL pagination, or writing OAuth refresh loops, your engineering team can focus on designing high-leverage workflows.

Whether you are building autonomous support agents that handle billing inquiries, or empowering your finance team to reconcile settlements via natural language queries, Truto provides the secure translation layer necessary to make it work reliably in production.