Skip to content
MCP server

pyeconomics

MCP server giving AI assistants cited economics and finance models with provenance

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, API, self-hosted
Runs
Locally
Version
1.0.0b1
Listing updated
today

Similar MCP servers

About pyeconomics

pyeconomics is an open-source MCP server for finance and payments. It runs locally on your machine and is started by your AI client.

The README describes pyeconomics as a Python library of 50 economics and finance model families across 13 domains, with a data layer covering 31 providers. The MCP server exposes that same model set to an AI assistant through five tools rather than one tool per model, so the assistant can find a model, read its inputs and run it.

Each run returns the same canonical JSON that the Python API produces, along with the manifest of every data source it read. Sources whose licence terms do not permit reaching a language model are filtered out before results leave the server, and an opt-in is required before unclear-terms data is passed through. Sixteen widely used models also get dedicated tools.

It suits developers and analysts who want an assistant to run valuation, fixed income, risk or portfolio calculations and cite the data behind them. Per the docs, it runs over stdio for desktop clients or as Streamable HTTP, and the README notes it is for educational and informational use, not investment advice.

Install

Run in a terminal

claude mcp add pyeconomics -- uvx pyeconomics

Add to claude_desktop_config.json

{
  "mcpServers": {
    "pyeconomics": {
      "command": "uvx",
      "args": [
        "pyeconomics"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "pyeconomics": {
      "command": "uvx",
      "args": [
        "pyeconomics"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "pyeconomics": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "pyeconomics"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "pyeconomics": {
      "command": "uvx",
      "args": [
        "pyeconomics"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "pyeconomics": {
      "command": "uvx",
      "args": [
        "pyeconomics"
      ]
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Tools this server provides (5)

Taken from the project documentation; the running server may differ.

search_models
Searches models by words in their ids, titles, tags, domains and summaries, or lists a domain's models.
describe_model
Returns a model's specification, input and output JSON Schemas and model card.
run_model
Runs any model on its inputs, optionally binding inputs to the data catalog.
get_series
Returns a catalog concept's or provider series' data with source records and attributions.
list_sources
Lists every data provider with licence class, LLM-use class and attribution terms.

What it does

Unified model tools
Five generic tools reach every registered model and data source, and sixteen common models have their own named tools.
Reproducible results
Each run returns canonical JSON matching the Python API along with a one-line summary.
Source provenance manifests
Results carry the manifest of each data source read, plus the attributions that must be shown.
Language-model data gating
Sources whose terms forbid use with a language model are skipped, and unclear sources need explicit opt-in.
Stdio and HTTP transports
Runs over standard input and output for desktop clients or as a Streamable HTTP server on localhost.

Good for

  • Calculating bond duration, yields or option Greeks inside an AI assistant
  • Running discounted cash flow or dividend valuation with cited inputs
  • Pulling macroeconomic or market series with their required attributions
  • Comparing portfolio risk metrics such as Sharpe ratio or value-at-risk
  • Checking which economic data sources are allowed for use with an assistant

Pricing

Open source

Apache License 2.0 from 1.0; the 1.0 beta is free to install from PyPI.

Prices change. Check pyeconomics.readthedocs.io before you buy.

pyeconomics FAQ

How do I install the server?

The docs give the command uvx --prerelease allow --from "pyeconomics[mcp]" pyeconomics mcp, or pip install --pre "pyeconomics[mcp]" followed by the same command.

Does the server run code or read files?

The docs say no tool executes code, fetches a URL you name, reads or writes a file, or returns an API key.

Is it stable?

The README says 1.0.0b1 is a beta pre-release with no stability promise until 1.0.0.

Can an assistant access data that forbids LLM use?

Forbidden sources are skipped and unclear ones need an opt-in set by the user in configuration, which the assistant cannot change.

Is pyeconomics free?

Yes. pyeconomics is open source, so you can use and self-host it for free.

How do I add pyeconomics to Claude Code?

Run this command in your terminal: claude mcp add pyeconomics -- uvx pyeconomics

What are the best alternatives to pyeconomics?

The closest alternatives to pyeconomics are QuantLedger 73-Engine CPA, GAAP, IRC Tax, Credit & SEC 10-K API, QuantLedger 73-Engine CPA, GAAP, IRC Tax, Credit & SEC 10-K API, meirinho.com.br, leitura para agentes, Llmmcp and awesome-mcp-servers.

Where this comes from

  • Description: written from the project's README, the project's website and its MCP Registry entry.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers