# BPstat Banco de Portugal AI Agent Connect

> BPstat is the official statistical database of Banco de Portugal. This MCP lets your AI client pull real-time interest rates, inflation metrics, balance of payments, and national financial accounts. You can navigate 77 statistical domains, pull specific time-series observations as CSV, or grab the latest reference values without needing any authentication.

## Overview
- **Category:** data-analytics
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_DyA58Ll4YT0nslnFwhgv2DQQfUqYkp7q2pz22Spo/ai-agent-connect
- **Tags:** portugal, central-bank, economic-statistics, interest-rates, monetary-data, time-series

## Description

You can now connect your AI agent directly to the primary source of Portuguese economic data. Instead of relying on second-hand aggregators, you use this MCP to query the Banco de Portugal database directly. It covers everything from the Lisbor interest rate and monetary aggregates to external reserves and payment systems. You can browse the entire domain tree to find specific series, inspect detailed metadata to understand dimensional breakdowns like sector or maturity, and pull historical time-series data. If you only need the most recent figure for a calculation or a report, you can skip the history and pull just the latest observation. It is a clean, zero-auth way to feed your agent the exact numbers used by economists and the ECB.

## Tools

### list_statistical_domains
This tool lists the 77 statistical domains available in BPstat, such as monetary, external, or interest rate data. Use the parent ID to navigate through sub-domains.

### list_domain_series
This tool retrieves the statistical series within a specific domain. It uses pagination to help you find the right series without overloading the request.

### get_series_metadata
This tool provides full metadata for a specific series, including its source, dimensional breakdowns, and the last recorded value.

### get_series_observations
This tool pulls historical time-series observations as a CSV. You can apply start and end date filters to keep the data payload manageable.

### get_latest_values
This tool returns the most recent reference date and value for a series. It is a lightweight way to get current figures like the latest inflation or interest rate.

## Prompt Examples

**Prompt:** 
```
What is the latest value of BPstat series 12456320 and what does it measure?
```

**Response:** 
```
Series 12456320 measures the net lending/net borrowing of the total economy through accumulated transactions. The latest value is 2.38 as of 2026-03-31. The historical range goes from a minimum of -12.5 during the 2012 crisis to a maximum of 6.9.
```

**Prompt:** 
```
Pull the 2025 monthly history of debt securities stock series 12834303
```

**Response:** 
```
The 2025 monthly history for series 12834303 (debt securities stock) is: Jan 523.94, Feb 527.93, Mar 527.05, and Dec 538.12. The data shows a steady monthly growth of 2.7% over the year.
```

**Prompt:** 
```
What statistical domains does Banco de Portugal publish and where are interest rates?
```

**Response:** 
```
BPstat publishes 77 domains including National financial accounts, External statistics, Payment systems, Cash issuance, and International reserves. Interest rates are located in their own dedicated domain, which includes Lisbor and retail bank rates.
```

## Capabilities

### Domain Navigation
Your agent uses this to crawl the 77 statistical categories to find specific data points.

### Metadata Inspection
Your agent checks this to understand the context, source, and dimensional members of a series.

### Time-Series Extraction
Your agent pulls historical data as CSV files for analysis or graphing.

### Real-time Updates
Your agent fetches the single most recent value for immediate reporting.

## Use Cases

### Economic Modeling
An agent pulls historical inflation and interest rate series to feed into a custom economic model.

### Market Monitoring
A user asks their agent to check the latest Lisbor rate to monitor liquidity conditions.

### Financial Reporting
A researcher uses the CSV output of debt securities stock to build a quarterly report.

### Data Verification
A journalist uses the metadata tool to confirm the source and methodology of a specific statistic.

## Benefits

- Accesses the same data used by the ECB and professional economists.
- Eliminates the need for manual API authentication or complex setup.
- Provides direct CSV output for time-series data.
- Reduces data errors by pulling from the authoritative central bank source.

## How It Works

Connecting to this MCP gives your AI client immediate access to the Banco de Portugal database.

1. Connect your MCP-compatible client to Vinkius.
2. Your agent explores the statistical domain tree to locate relevant data.
3. The agent identifies specific series IDs using the list tools.
4. The agent pulls the requested data as JSON or CSV.
5. The data is ready for your agent to analyze or present.

## Frequently Asked Questions

**Do I need an API key to use BPstat data?**
No, this MCP uses zero-auth access to the Banco de Portugal statistical database.

**Can I get the latest interest rates?**
Yes, you can use the get_latest_values tool to pull the most recent reference date and value for interest rate series.

**What kind of data is available?**
The database includes monetary aggregates, inflation, interest rates, balance of payments, external reserves, and national financial accounts.

**Which AI clients can use this MCP?**
You can use this with any MCP-compatible client, including Claude, Cursor, Windsurf, and VS Code.

**How can I filter the historical data I pull?**
You can use the start_date and end_date parameters in the get_series_observations tool to limit the results to a specific timeframe.

**Do I need an API key or registration?**
No. BPstat's data API is fully public — every tool in this MCP works without authentication or account.

**How do I find the series id I need?**
Start with list_statistical_domains, then list_domain_series for the domain you care about. Each series entry includes its id, title and last value. You can also browse bpstat.bportugal.pt/pt/dados/explorer in a browser and copy the series id from the URL.