MCP server acting as an interface to the Frankfurter API for currency exchange data.
Frankfurter is a useful API for latest currency exchange rates, historical data, or time series published by sources such as the European Central Bank. Should you have to access the Frankfurter API as tools for language model agents exposed over the Model Context Protocol (MCP), Frankfurter MCP is what you need.
Following is a table of some updates regarding the project status. Note that these do not correspond to specific commits or milestones.
Date | Status | Notes or observations |
---|---|---|
June 13, 2025 | active | Added LlamaIndex tool listing for demonstration only. (The --all-extras flag is necessary to install LlamaIndex, which is not installed by default.) |
June 9, 2025 | active | Added containerisation, support for self-signed, proxies. |
June 8, 2025 | active | Added dynamic composition. |
June 7, 2025 | active | Added tools to cover all the functionalities of the Frankfurter API. |
June 7, 2025 | active | Project started. |
The directory where you clone this repository will be referred to as the working directory or WD hereinafter.
Install uv. To install the project with its minimal dependencies in a virtual environment, run the following in the WD. To install all non-essential dependencies (which are required for developing and testing), replace the --no-dev
with the --all-groups
flag in the following command.
uv sync --no-dev
Following is a list of environment variables that can be used to configure the application. A template of environment variables is provided in the file .env.template
.
The following environment variables can be specified, prefixed with FASTMCP_
: HOST
, PORT
, DEBUG
and LOG_LEVEL
. See global configuration options for FastMCP. Note that on_duplicate_
prefixed options specified as environment variables will be ignored.
The underlying HTTP client also respects some environment variables, as documented in the HTTPX library. In addition, SSL_CERT_FILE
and SSL_CERT_DIR
can be configured to so that
Variable | [Default value] and description |
---|---|
HTTPX_TIMEOUT | [5.0] The time for the underlying HTTP client to wait, in seconds, for a response. |
HTTPX_VERIFY_SSL | [True] This variable can be set to False to turn off SSL certificate verification, if, for instance, you are using a proxy server with a self-signed certificate. However, setting this to False is advised against: instead, use the SSL_CERT_FILE and SSL_CERT_DIR variables to properly configure self-signed certificates. |
FAST_MCP_HOST | [0.0.0.0] This variable specifies which host the MCP server must bind to unless the server transport (see below) is set to stdio . |
FAST_MCP_PORT | [8000] This variable specifies which port the MCP server must listen on unless the server transport (see below) is set to stdio . |
MCP_SERVER_TRANSPORT | [streamable-http] The acceptable options are stdio , sse or streamable-http . Given the use-case of running this MCP server as a remotely accessible endpoint, there is no real reason to choose stdio . |
MCP_SERVER_INCLUDE_METADATA_IN_RESPONSE | [True] An experimental feature to include additional metadata to the MCP type TextContent that wraps the response data from each tool call. The additional metadata, for example, will include (as of June 21, 2025) the API URL of the Frankfurter server that is used to obtain the responses. |
FRANKFURTER_API_URL | [https://api.frankfurter.dev/v1] If you are self-hosting the Frankfurter API, you should change this to the API endpoint address of your deployment. |
pip
)Add this package from PyPI using pip
in a virtual environment (possibly managed by conda
or pyenv
) and then start the server by running the following.
Add a .env
file with the contents of the .env.template
file if you wish to modify the default values of the aforementioned environment variables. Or, on your shell, you can export the environment variables that you wish to modify.
pip install frankfurtermcp
python -m frankfurtermcp.server
uv
)Copy the .env.template
file to a .env
file in the WD, to modify the aforementioned environment variables, if you want to use anything other than the default settings. Or, on your shell, you can export the environment variables that you wish to modify.
Run the following in the WD to start the MCP server.
uv run frankfurtermcp
If you want to run it without uv
, assuming that the appropriate virtual environment has been created in the .venv
within the WD, you can start the server calling the following.
./.venv/bin/python -m frankfurtermcp.server
The MCP endpoint will be available over HTTP at http://localhost:8000/sse for the Server Sent Events (SSE) transport, or http://localhost:8000/mcp for the streamable HTTP transport. To exit the server, use the Ctrl+C key combination.
There is one Dockerfile provided in this repository.
local.dockerfile
for using the latest version, which can contain your edits to the code if you do make edits.To build the image, create the container and start it, run the following in WD. Choose shorter names for the image and container if you prefer.
If you change the port to anything other than 8000 in .env.template
, do remember to change the port number references in the following command. Instead of passing all the environment variables using the --env-file
option, you can also pass individual environment variables using the -e
option.
docker build -f local.dockerfile -t frankfurtermcp .
docker create -p 8000:8000/tcp --env-file .env.template --expose 8000 --name frankfurtermcp-container frankfurtermcp
docker start frankfurtermcp-container
Upon successful build and container start, the MCP server will be available over HTTP at http://localhost:8000/sse for the Server Sent Events (SSE) transport, or http://localhost:8000/mcp for the streamable HTTP transport.
To see how to use the MCP server by mounting it dynamically with FastMCP, check the file src/frankfurtermcp/composition.py
.
The following table lists the names of the tools as exposed by the FrankfurterMCP server. It does not list the tool(s) exposed through the composition example. The descriptions shown here are for documentation purposes, which may differ from the actual descriptions exposed over the model context protocol.
Name | Description |
---|---|
get_supported_currencies | Get a list of currencies supported by the Frankfurter API. |
get_latest_exchange_rates | Get latest exchange rates in specific currencies for a given base currency. |
convert_currency_latest | Convert an amount from one currency to another using the latest exchange rates. |
get_historical_exchange_rates | Get historical exchange rates for a specific date or date range in specific currencies for a given base currency. |
convert_currency_specific_date | Convert an amount from one currency to another using the exchange rates for a specific date. |
The required and optional arguments for each tool are not listed in the following table for brevity but are available to the MCP client over the protocol.
If you want to see the detailed schema for a particular tool, you can do so using the tools-info
commmand from the command line interface. The command line interface is available as the script cli
. You can invoke its help to see the available commands as follows.
uv run cli --help
This will produce an output similar to the screenshot below.
Before calling the tools-info
command, you MUST have the server running in streamable-http
or sse
transport mode, preferably locally, e.g., by invoking uv run frankfurtermcp
. A successful call of the tools-info
command will generate an output similar to the screenshot shown below.
Alternative to the tools-info
command, you can also run call the llamaindex-tools-list
command to display the names of the tools without the respective function schemas. This functionality is provided only to optionally demonstrate that the LlamaIndex MCP client can read the tools list from this MCP server. In order for this to function, you must install LlamaIndex MCP client by calling uv sync --extra opt
. The output of calling this command will look like the following.
Install pre-commit
for Git and ruff
. Then enable pre-commit
by running the following in the WD.
pre-commit install
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
To run the provided test cases, execute the following. Add the flag --capture=tee-sys
to the command to display further console output.
Note that for the tests to succeed, the environment variable MCP_SERVER_TRANSPORT
must be set to either sse
or streamable-http
, or not set at all, in which case it will default to streamable-http
.
uv run --group test pytest tests/
MIT.
Knowledge graph-based persistent memory system
MCP Server For Apache Doris, an MPP-based real-time data warehouse.
Official MCP Server from Atlan which enables you to bring the power of metadata to your AI tools
Query Onchain data, like ERC20 tokens, transaction history, smart contract state.
Read and write access to your Baserow tables.
Embeddings, vector search, document storage, and full-text search with the open-source AI application database
Query your ClickHouse database server.
Official CoinGecko API MCP Server for Crypto Price & Market Data, across 200+ blokchain networks and 8M+ tokens.
Introspect and query your apps deployed to Convex.
Access comprehensive B2B data on companies, employees, and job postings for your LLMs and AI workflows.