BCRP MCP Server
Enables AI agents to access economic statistics from the Central Reserve Bank of Peru, including exchange rates, interest rates, and GDP, without requiring an API key.
README
BCRP MCP Server
Connects any MCP-compatible AI agent to BCRPData — the official open-access economic statistics database of the Banco Central de Reserva del Perú (BCRP).
Access real-time and historical data for exchange rates, interest rates, international reserves, monetary liquidity, private sector credit, commodity prices, stock market indicators, and GDP — no API key required.
Works with: Claude Desktop · Claude Code · Cursor · VS Code Copilot · Windsurf · Zed · Continue.dev · any MCP client
Quickstart (2 minutes)
1. Install the server
# Option A: uvx (recommended — zero setup)
uvx bcrp-mcp
# Option B: pip
pip install bcrp-mcp
No API key needed. BCRPData is a fully public API.
2. Add to your AI client — see the Configuration section.
3. Start using it:
> What is the current BCRP policy rate and exchange rate?
> Use bcrp_get_macro_snapshot to get a snapshot of Peru's key economic indicators.
> Show me the evolution of Peru's international reserves since 2020.
Prerequisites
Python 3.11+
python --version # needs 3.11 or higher
uv (recommended)
# Windows
winget install astral-sh.uv
# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
Installation
Option A — uvx (zero configuration)
uvx bcrp-mcp
# Upgrade later
uv tool upgrade bcrp-mcp
Option B — pip
pip install bcrp-mcp
bcrp-mcp # starts the server
Option C — From source (development)
git clone https://github.com/JOSETRA44/bcrp-mcp.git
cd bcrp-mcp
uv sync
cp .env.example .env # optional — customize settings
uv run bcrp-mcp
Configuration (Optional)
All settings have sensible defaults. Override via environment variables or .env file:
| Variable | Default | Description |
|---|---|---|
BCRP_CACHE_TTL |
300 |
Response cache in seconds (0 = disabled) |
BCRP_TIMEOUT |
30 |
HTTP timeout in seconds |
BCRP_MAX_RETRIES |
3 |
Retries on transient errors |
BCRP_LANGUAGE |
esp |
Default language: esp (Spanish) or ing (English) |
LOG_LEVEL |
INFO |
DEBUG · INFO · WARNING · ERROR |
No API key required — BCRPData is fully open access.
Configuration by Client
Claude Desktop
Config file:
- Windows:
%APPDATA%\Claude\claude_desktop_config.json - macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"bcrp": {
"command": "uvx",
"args": ["bcrp-mcp"]
}
}
}
Restart Claude Desktop after saving. You'll see a hammer icon (🔨) confirming the server is loaded.
Claude Code (CLI)
Add to your .mcp.json or .antigravity.json:
{
"mcpServers": {
"bcrp": {
"command": "uvx",
"args": ["bcrp-mcp"]
}
}
}
From source:
{
"mcpServers": {
"bcrp": {
"command": "uv",
"args": [
"--directory", "/absolute/path/to/bcrp-mcp",
"run", "bcrp-mcp"
]
}
}
}
Run /mcp in the Claude Code prompt to confirm — you should see bcrp with 6 tools.
Cursor
{
"mcpServers": {
"bcrp": {
"command": "uvx",
"args": ["bcrp-mcp"]
}
}
}
VS Code + GitHub Copilot
Create .vscode/mcp.json:
{
"servers": {
"bcrp": {
"type": "stdio",
"command": "uvx",
"args": ["bcrp-mcp"]
}
}
}
Windsurf
Config: %APPDATA%\Codeium\windsurf\mcp_config.json
{
"mcpServers": {
"bcrp": {
"command": "uvx",
"args": ["bcrp-mcp"]
}
}
}
Zed
Edit ~/.config/zed/settings.json:
{
"context_servers": {
"bcrp": {
"command": {
"path": "uvx",
"args": ["bcrp-mcp"]
}
}
}
}
Continue.dev
Edit .continue/config.yaml:
mcpServers:
- name: bcrp
command: uvx
args:
- bcrp-mcp
Generic stdio
command: uvx
args: ["bcrp-mcp"]
Available Tools (6)
| Tool | Description |
|---|---|
bcrp_get_series |
Fetch 1–10 series with optional date range (all same frequency) |
bcrp_get_latest |
Get the most recent N data points for 1–10 series |
bcrp_describe_series |
Metadata only: series names, frequency, available date range |
bcrp_search_catalog |
Search built-in catalog of important series by keyword |
bcrp_list_catalog_categories |
List all catalog categories with series counts |
bcrp_get_macro_snapshot |
Real-time snapshot of Peru's key macroeconomic indicators |
Available Prompts (3)
| Prompt | Arguments | Description |
|---|---|---|
inflation_and_monetary_analysis |
start_year, end_year |
Structured workflow: policy rate, liquidity (M2), credit growth |
exchange_rate_analysis |
start_date, end_date |
Exchange rate dynamics, reserves, dollarization |
economic_growth_analysis |
start_year, end_year |
GDP growth, monetary multiplier, credit cycle |
Available Resources (3)
| Resource URI | Contents |
|---|---|
bcrp://api-guide |
Complete API reference: URL structure, period formats, response schema |
bcrp://period-formats |
Quick reference table for all period formats by frequency |
bcrp://series-catalog |
Curated catalog of important series with codes, names, units |
Key Series Codes
Daily (format: DD-MM-YYYY)
| Code | Indicator | Unit |
|---|---|---|
PD04637PD |
TC Interbancario Compra (USD/PEN) | S/ por US$ |
PD04638PD |
TC Interbancario Venta (USD/PEN) | S/ por US$ |
PD12301MD |
Tasa de Referencia BCRP | % anual |
PD04692MD |
Tasa Interbancaria, S/ | % anual |
PD04650MD |
Reservas Internacionales Netas | Millones US$ |
PD38026MD |
Índice General Bursátil BVL | Índice |
PD04701XD |
Precio del Cobre | cUS$ por libra |
PD04705XD |
Precio del Petróleo WTI | US$ por barril |
PD04721XD |
Índice Dow Jones | Índice |
Monthly (format: YYYY-M)
| Code | Indicator | Unit |
|---|---|---|
PN00196MM |
Liquidez Total (M2) | Millones S/ |
PN00178MM |
Circulante (M0) | Millones S/ |
PN00190MM |
Liquidez en Soles | Millones S/ |
PN00180MM |
Liquidez var% 12 meses | % anual |
PN00496MM |
Crédito Sector Privado | Millones S/ |
PN00500MM |
Crédito var% 12 meses | % anual |
Quarterly (format: YYYY-Q)
| Code | Indicator | Unit |
|---|---|---|
PN03503MQ |
PBI Nominal var% | % trimestral |
PN03501MQ |
Liquidez MN var% | % trimestral |
PN03502MQ |
Emisión Primaria var% | % trimestral |
PN03504MQ |
Multiplicador Monetario var% | % trimestral |
Example Queries
# Get the current exchange rate (last 5 trading days)
bcrp_get_latest(series_codes=["PD04637PD", "PD04638PD"], n_periods=5)
# BCRP policy rate since 2022
bcrp_get_series(
series_codes=["PD12301MD"],
start_period="01-01-2022",
end_period="30-06-2026"
)
# Monthly liquidity (M2) and credit growth last 2 years
bcrp_get_series(
series_codes=["PN00196MM", "PN00496MM", "PN00500MM"],
start_period="2024-1",
end_period="2026-6"
)
# Complete macro snapshot
bcrp_get_macro_snapshot()
# Search for more series
bcrp_search_catalog(query="tipo de cambio", frequency="monthly")
bcrp_search_catalog(query="exportaciones")
API Notes
- No authentication required — BCRPData is fully public.
- Multiple series in one call must be the same frequency (all daily, all monthly, etc.).
- Period formats differ by frequency — see
bcrp://period-formatsresource. - The BCRP database has 8,000+ monthly, 2,700+ quarterly, and 800+ daily series. Browse the full catalog at estadisticas.bcrp.gob.pe.
Verify It's Working
# Interactive browser UI
npx @modelcontextprotocol/inspector uvx bcrp-mcp
# Quick smoke test
echo "" | uvx bcrp-mcp
# Unit tests (from source)
uv sync --group dev
uv run pytest tests/ -v
Troubleshooting
command not found: uvx
Install uv: https://docs.astral.sh/uv/getting-started/installation/
Empty results for a series
The series code may be discontinued or the period range may not have data.
Run bcrp_describe_series first to check available date range.
Series codes with mismatched frequencies All series in a single call must share the same frequency. Daily and monthly series cannot be mixed — make two separate calls.
Slow first start uv downloads and caches the package on first run. Subsequent starts take ~0.2s.
Project Structure
bcrp-mcp/
├── src/bcrp_mcp/
│ ├── server.py # FastMCP entry point
│ ├── config.py # Settings (pydantic-settings, no auth required)
│ ├── client.py # Async HTTP client + TTL cache + retry
│ ├── exceptions.py # Error hierarchy
│ ├── formatters.py # Raw BCRP JSON → clean AI-friendly dicts
│ ├── catalog.py # Curated series catalog + search
│ ├── tools/ # 6 MCP tools
│ ├── prompts/ # 3 MCP prompts (analysis workflows)
│ └── resources/ # 3 MCP resources (guides + catalog)
├── tests/ # Unit tests (no network required)
├── .env.example # Optional configuration
└── pyproject.toml
License
MIT — © 2026 JOSETRA44
Data Source
All data comes from BCRPData — the official open statistics platform of the Banco Central de Reserva del Perú. For terms of use, see: https://estadisticas.bcrp.gob.pe/estadisticas/series/ayuda/condicionesUso
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.