Does RateAPI cover more than lending rates?

Yes. In addition to mortgages, auto loans, HELOCs, personal loans, and credit cards, the MCP includes deposit and CD benchmarks, credit union profiles, NCUA facts, branch discovery, membership eligibility, calculations, monitoring, portfolio scans, reports, alerts, and campaign workflows.

AI Agents

Build an AI Agent That Finds the Better Financial Deal

Let Claude, ChatGPT, VS Code, and other agents verify a quoted rate, search independent alternatives, confirm eligibility, calculate the real cost, and guide the next move. One MCP server provides the 74 tools behind the answer.

MCP Setup Guide API Documentation

Last updated: August 4, 2026

Quick Answer

What does RateAPI add to an AI agent?

RateAPI turns a general-purpose model into a source-backed financial research and decision agent. It can retrieve current loan and deposit rates, find and inspect credit unions, check whether a person can join, locate nearby branches, run financing math, watch markets, and produce client-ready outputs—without pretending training data is live data.

74 tools: research, decisions, calculations, monitors, portfolios, reports, alerts, and campaigns.
Full market surface: mortgages, auto loans, HELOCs, personal loans, credit cards, savings, money markets, and CDs.
Evidence attached: source URLs, verification timestamps, membership rules, NCUA facts, and transparent ranking logic.
Connect in one command via the RateAPI MCP server, or use REST and function calling.
Free tier covers 20 requests/month (email required) for testing—see pricing.

The Problem

A Language Model Is Not a Live Market, a Calculator, or a Fiduciary

A model can explain APR, but it cannot know today's offers from thousands of institutions unless it retrieves them. It also cannot infer that a borrower qualifies for a credit union, that a branch is nearby, or that a refinance breaks even in 19 months without the right data and tools.

Sending someone to "ask a lender" is not neutral price discovery. A lender can explain its own offer; it is not searching the entire market for the lowest-cost option. An agent needs an independent data layer before it can help a user compare rather than simply sell.

Your AI agent can do the work. It can retrieve live offers, compare true cost, verify the institution, test membership eligibility, find a branch, calculate the tradeoff, and cite the evidence in the same conversational turn.

RateAPI does not sell leads or accept paid lender placement. The agent gets a market view ranked by borrower cost—not a list ordered by who paid for the click.

Integration

Two Ways to Connect

🤖

Remote MCP Server

Zero integration code

Add the remote MCP server to Claude, ChatGPT, VS Code, or another compatible client and it can discover the RateAPI tool catalog automatically. Just add the server and start asking.

$bash

# Add to Claude Code
claude mcp add rateapi \
  https://mcp.rateapi.dev/mcp \
  --transport http \
  --header "x-api-key: YOUR_KEY"

🛠

Function Calling API

Works with any LLM

Use the OpenAPI spec to generate tool schemas for your own application. The same underlying data and decisions are available through standard REST endpoints.

{}json

// OpenAI function call response
{
  "name": "get_financing_decision",
  "arguments": {
    "product_type": "auto_loan",
    "state": "CA",
    "loan_amount": 32000,
    "term_months": 60
  }
}

Features

A Complete Financial Toolbelt

🔌

Rates and Financing Decisions

Compare mortgages, auto loans, HELOCs, personal loans, and credit cards. Rank offers, explain the reasoning, and calculate savings and total cost.

🎯

Deposits and Market Intelligence

Research savings, money-market, CD, and IRA CD APYs. Pull benchmarks, rate history, competitive position, and institution league tables.

💬

Credit Union Research

Retrieve source-labelled profiles, NCUA member and asset facts, services, branch locations, current rates, and data freshness in one workflow.

🔔

Eligibility and Nearby Branches

Check field-of-membership rules using geography, employer, military service, school, or worship—then find the nearest reported branch by ZIP.

⚖

Calculators and Guardrails

Run affordability, refinance break-even, buydown, amortization, PMI, rent-vs-buy, and rate-lock analysis with advice guardrails.

⚙

Ongoing Agent Workflows

Save scenarios, monitor rates, scan portfolios, schedule digests, generate client reports, and manage branded rate-email campaigns.

Examples

Conversations Your Agent Can Handle

Eligibility + Branch + Auto Financing

User

I'm buying a used car near 95814. Find a competitive 60-month loan from a credit union I can actually join.

Agent

I'll combine four tools: current used-auto offers, true-cost ranking, membership eligibility, and nearest-branch discovery. I'll label any uncertain eligibility instead of hiding the institution, and include the credit union's membership rule and source link with each recommendation.

Cross-Product Financial Decision

User

I have $50,000 in savings and a mortgage at 7.25%. Should I move the cash to a CD, pay down the loan, or refinance?

Agent

I'll compare current CD and money-market APYs against the after-tax value of reducing your mortgage balance, then run a refinance break-even using live offers and closing-cost assumptions. I can save the scenario and create a monitor so the recommendation updates when the math changes.

Resources

Developer Resources

📄

OpenAPI Spec

Full API reference with schemas for tool generation

🤖

MCP Setup Guide

Setup for Claude, ChatGPT, VS Code, and other MCP clients

💻

Example Prompts

Sample system prompts and tool definitions

Try the MCP Server Now

Add the remote MCP server to a compatible client and start with a real task: compare financing, research a credit union, check eligibility, analyze deposits, or create a rate monitor. Setup takes about two minutes.

MCP Setup Guide MCP Server Docs

FAQ

Frequently Asked Questions

What can an AI agent do with RateAPI?

RateAPI exposes 74 tools covering live lending and deposit rates, financing decisions, credit union and branch discovery, membership eligibility, affordability and refinance math, benchmarks, monitors, portfolios, reports, alerts, and email campaigns.

What is MCP (Model Context Protocol)?

MCP (Model Context Protocol) is an open standard for connecting AI applications to tools and data. RateAPI provides one remote server for Claude, ChatGPT, VS Code, and other compatible clients. Add the server once and the client discovers the available tools automatically.

Can I use RateAPI with ChatGPT?

Yes. ChatGPT can use the remote MCP server in MCP-compatible surfaces. Developers can also use RateAPI's REST API with OpenAI function calling when building a custom application.

How accurate is the rate data?

RateAPI collects rates directly from 3,700+ credit union websites and returns verification timestamps and source evidence. Unlike lead-generation marketplaces, RateAPI does not accept paid lender placement or rank by referral revenue.

Is there a free tier for AI agents?

Yes. The free tier includes 20 requests per month and is designed for testing an agent integration. Pro starts at $49/month, with enterprise options for larger automated workflows.

How do I stop my AI agent from hallucinating financial rates?

Give it a live retrieval tool instead of asking it to remember a changing market. RateAPI returns current data, source links, and verification timestamps at query time, so the model can cite retrieved facts and keep its own reasoning separate.

Does the MCP cover more than lending rates?

Yes. It covers auto loans, HELOCs, personal loans, and credit cards, plus savings, money-market, CD, and IRA CD data. It also includes NCUA profiles, eligibility, branch discovery, calculations, monitoring, portfolio, reporting, and campaign tools.

Ready to Give Your AI a Financial Toolbelt?

Connect one MCP server and let your agent research, calculate, verify, monitor, and act with live financial data. The free tier includes 20 requests/month.

MCP Setup Guide