---
title: "Connect an AI agent with the MCP server"
description: "Wire Claude, Cursor, or any MCP-compatible host to the BlindPay API with the official MCP server and agent skills, with guardrails for money movement."
date: "2026-08-26"
category: "integrations"
products: ["payouts", "payins", "compliance"]
---

Paste this prompt into your coding agent to give an AI assistant safe, tool-based access to BlindPay: balances, customers, quotes, and payments through natural language.

## Prompt

```text
You are connecting an AI agent (Claude Desktop, Claude Code, Cursor, or another MCP-compatible host) to BlindPay using the official MCP server, with guardrails appropriate for tools that move money.

Before configuring anything, read these sources and follow them over any prior knowledge:
- https://blindpay.com/docs/llms.txt (read the build with AI page)
- The MCP server card: https://blindpay.com/.well-known/mcp/server-card.json

Set up:
1. Install the MCP server in my host's config: command npx, args ["-y", "@blindpay/mcp"], env BLINDPAY_API_KEY and BLINDPAY_INSTANCE_ID. Use a DEVELOPMENT instance ID and key first; document exactly which config file this lands in for my host.
2. Add the agent skills knowledge layer: npx skills add blindpaylabs/skills, so the agent knows rails, corridors, fees, and KYC patterns instead of guessing.
3. Verify the connection: have the agent list its BlindPay tools, then run a read-only call (for example listing customers) before anything that writes.
4. Guardrails, written into my agent's system prompt or host permissions:
   - Read operations are fine unprompted; any operation that creates a payment (quotes are fine, payouts/payins/transfers are not) requires explicit human confirmation with the amount and destination restated.
   - The agent must never echo the API key, and must refuse to switch to a production instance ID without me changing the config myself.
   - On a development instance, funds are test-only (USDB on testnets); state this in responses so nobody mistakes a test payout for a real one.
5. Produce a short runbook: how to rotate the key (delete and recreate in the dashboard; keys are shown once), how to point at production later, and which operations the agent is allowed to run at each stage.

Deliverables: the host config snippet, the skills install, a verified tool-listing transcript, the guardrail prompt text, and the runbook.
```

## How to use

1. Create a development instance and API key at [app.blindpay.com](https://app.blindpay.com) before starting.
2. Paste the prompt and name your MCP host; config file locations differ per host.
3. Keep the guardrails when you move to production; raise the confirmation bar rather than lowering it.

## Related docs

- [Build with AI](https://blindpay.com/docs/build-with-ai)
- [BlindPay for AI](https://blindpay.com/ai)
- [API keys](https://blindpay.com/docs/learn/api-keys)