About IMF Data MCP
IMF Data MCP connects clients to economic datasets from the IMF’s current data.imf.org SDMX API. It lets analysts and researchers discover databases and dimensions, then retrieve filtered time series.
Key features
- 01List and search IMF databases
- 02Inspect query dimensions and valid codes
- 03Fetch filtered time series
- 04Apply automatic rate limiting and retries
Use cases
- 01Find databases for an economic topic
- 02Retrieve country-specific indicator data
- 03Analyze time series across a year range
MCP support
The server exposes IMF database discovery and data retrieval as MCP tools. It also provides IMF data query capabilities to MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEIMF Data MCP
A Model Context Protocol server for IMF economic data, built on
imfp and the
IMF SDMX API at data.imf.org.
v0.2.0 — Breaking change: The old server called
dataservices.imf.org, which the IMF decommissioned in 2025. All requests timed out as a result. This version migrates toimfp, which targets the currentdata.imf.orgSDMX 3.0 API and is actively maintained.
Installation
Using uv (recommended)
uvx --from git+https://github.com/c-cf/imf-data-mcp imf-data-mcp
Using pip
pip install git+https://github.com/c-cf/imf-data-mcp.git
Claude Desktop configuration
{
"mcpServers": {
"imf": {
"command": "uvx",
"args": ["--from", "git+https://github.com/c-cf/imf-data-mcp", "imf-data-mcp"]
}
}
}
Or, if installed via pip:
{
"mcpServers": {
"imf": {
"command": "imf-data-mcp"
}
}
}
Available tools
| Tool | Description |
|---|---|
imf_list_databases | List all ~155 IMF databases with their IDs |
imf_search_databases | Filter databases by keyword (e.g. "inflation", "trade") |
imf_get_parameter_defs | List query dimensions for a database |
imf_get_parameter_codes | List valid codes for a dimension, with optional search |
imf_fetch_data | Fetch time series data with year range and dimension filters |
Typical workflow
1. imf_search_databases(keyword="consumer price")
→ CPI, CPI_WCA, ...
2. imf_get_parameter_defs(database_id="CPI")
→ country, index_type, coicop_1999, frequency, ...
3. imf_get_parameter_codes(database_id="CPI", parameter="country", search="austria")
→ AUT
4. imf_fetch_data(
database_id="CPI",
start_year=2015, end_year=2023,
filters={"country": ["AUT"], "frequency": ["M"]}
)
Common database IDs
| Topic | ID |
|---|---|
| Consumer Prices | CPI |
| Producer Prices | PPI |
| National Accounts (Annual) | ANEA |
| National Accounts (Quarterly) | QNEA |
| Balance of Payments | BOP_AGG |
| Exchange Rates | ER_2026_JAN_VINTAGE |
| Commodity Prices | PCPS |
| Fiscal Monitor | FM |
| Gov. Finance Statistics | GFS_COFOG |
| Monetary Statistics | MFS_CBS, MFS_MA, MFS_IR |
| Financial Soundness | FSIC |
| Trade in Services | ITS |
| SDG Data | SDG |
Rate limits
The IMF API allows roughly 10 requests / 5 seconds. imfp handles
rate limiting and retries automatically.
Debugging
npx @modelcontextprotocol/inspector imf-data-mcp
License
Apache 2.0
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.