pyeconomics — Claude Code MCP server: what it does, how to install
pyeconomics — an MCP server with 50 families of economics and finance models: fixed income, derivatives, portfolios, risk, econometrics, macro and asset pricing. Every result ships with a provenance manifest.
Apache-2.0 ★ 11 updated 2026-10-29
claude mcp add pyeconomics -- uvx pyeconomicsWhat pyeconomics does
The server covers 13 domains: finance foundations, fixed income, derivatives, international finance, equity, corporate finance, accounting, portfolios, performance measurement, risk management, econometrics, macroeconomics and asset pricing. The data platform connects 31 providers and 171 catalog concepts with real-time vintages. Typed inputs declare units and bounds; rates and percentages are decimals — 0.05 stands for 5%.
Each model runs through the Python API, the pyeconomics CLI, the REST API (pyeconomics serve) or the MCP server (pyeconomics mcp). Requires Python 3.12 or later; beta version 1.0.0b1 needs explicit installation: pip install --pre pyeconomics==1.0.0b1. To run the server with MCP support, install the extra package: pip install --pre "pyeconomics[mcp]==1.0.0b1" or launch directly with uvx --from "pyeconomics[mcp]==1.0.0b1" pyeconomics mcp.
Results export to tables or reproducibility bundles. The stable 0.2.6 release includes four monetary policy rules over FRED data and remains MIT-licensed; the new version moves to Apache 2.0.
What it helps with
- price a bond or derivative with specified parameters
- compute portfolio metrics and risk profile
- run an econometric model on live data
- get a result with full provenance manifest
- export a calculation to a reproducible bundle
Example prompts for the agent
- Calculate portfolio return with 0.6 equity weight and 0.4 bonds
- Run an option pricing model with volatility 0.25 and strike 100
- Show the provenance manifest for the last calculation
How to install pyeconomics
Platform CLI
claude mcp add pyeconomics -- uvx pyeconomicsuvx runs the command named after the package; if the README names another one, use uvx --from pyeconomics <command>
Platform CLI
codex mcp add pyeconomics -- uvx pyeconomicsuvx runs the command named after the package; if the README names another one, use uvx --from pyeconomics <command>
Config file
{
"mcpServers": {
"pyeconomics": {
"command": "uvx",
"args": [
"pyeconomics"
]
}
}
}uvx runs the command named after the package; if the README names another one, use uvx --from pyeconomics <command>
Config file
{
"servers": {
"pyeconomics": {
"type": "stdio",
"command": "uvx",
"args": [
"pyeconomics"
]
}
}
}uvx runs the command named after the package; if the README names another one, use uvx --from pyeconomics <command>
Platform CLI
gemini mcp add pyeconomics uvx pyeconomicsuvx runs the command named after the package; if the README names another one, use uvx --from pyeconomics <command>
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about pyeconomics
How to install pyeconomics in Claude Code?
Run claude mcp add pyeconomics -- uvx pyeconomics.
What does pyeconomics do?
Provides access to 50 families of economics and finance models — fixed income, derivatives, portfolios, risk, econometrics, macro and asset pricing — through MCP. Every result includes a provenance manifest with formulas and assumptions.
Which Python version does pyeconomics require?
Python 3.12 or later. Beta version 1.0.0b1 requires explicit install: pip install --pre pyeconomics==1.0.0b1.