Open Finance US MCP
A standalone MCP server wrapping Mastercard Open Finance US (Finicity) APIs for open banking operations, enabling financial data access and payment services.
README
Open Finance US MCP
A standalone Model Context Protocol server
that wraps the Mastercard Open Finance US (Finicity) hosted APIs
(https://api.finicity.com). It is a separate codebase from Vima and vendors
its own Finicity client — no dependency on Vima's apis/, execute() wrapper,
or simulator.

Layout
openfinance-mcp/
src/openfinance_mcp/
finicity_client.py # vendored Finicity client (only depends on requests)
server.py # FastMCP server + tools
models.py # ToolResult envelope
spec_defaults.py # resolved spec decisions + open questions
config/settings.py # credential loader (env / dotenv / sibling vima config/.env)
__main__.py # stdio + streamable-http entry point
tests/ # unit, protocol (offline) + sandbox (live) suites
console/ # separate, independently deployable demo UI
run.sh / run.bat # one-shot start scripts for the MCP server
console/run.sh / .bat # one-shot start scripts for the Test Console
1 — Get Credentials
Credentials are issued through the Mastercard Developers portal.
-
Sign up or log in at
developer.mastercard.com. -
Click Create New Project, enter a project name, and select Open Finance from the Select your API service dropdown.
-
Select the commercial country for your end users, then click Proceed.
-
Enter a description, click Create Project.
-
In the left-hand Credentials panel click Sandbox — your three credential values are shown:
Variable Portal label OPEN_FINANCE_PARTNER_IDPartner ID OPEN_FINANCE_PARTNER_SECRETSecret OPEN_FINANCE_APP_KEYApp Key
Full quick-start guide: https://developer.mastercard.com/open-banking-us/documentation/quick-start-guide/
API base URL defaults to
https://api.finicity.com. Override withOPEN_FINANCE_API_BASE_URLif needed.
2 — Configure Credentials
Option A — Test Console UI (recommended)
Start the console (see below), then click the ⚙ Credentials button in the
top-right corner. Paste your values and click Save to .env. The file is
written to .env in the repo root.
Option B — Edit .env directly
Create .env in the repo root:
OPEN_FINANCE_PARTNER_ID=your_partner_id
OPEN_FINANCE_PARTNER_SECRET=your_partner_secret
OPEN_FINANCE_APP_KEY=your_app_key
OPEN_FINANCE_API_BASE_URL=https://api.finicity.com
Resolution order
- Process environment variables.
OPENFINANCE_MCP_ENV_FILE(explicit dotenv path)..envin this repo root.- Sibling
../vima/config/.env(auto-discovered).
3 — Start the MCP Server
The run scripts create a .venv, install dependencies, kill any previous
instance, and start the server — no manual pip install required.
macOS / Linux
./run.sh
# Defaults: transport=http, host=0.0.0.0, port=9030
Windows
run.bat
Override defaults via environment variables:
PORT=9030 TRANSPORT=http ./run.sh
Manual (without the script)
python -m venv .venv && source .venv/bin/activate
pip install -e ".[test,console]"
python -m openfinance_mcp --transport http --port 9030
4 — Start the Test Console
macOS / Linux
./console/run.sh
# open http://localhost:8080
Windows
console\run.bat
Override defaults:
PORT=8080 MCP_URL=http://localhost:9030/mcp ./console/run.sh
Manual
cd console
pip install -r requirements.txt
MCP_URL=http://localhost:9030/mcp uvicorn app:app --port 8080
Or start both MCP server and console together with Docker:
docker compose up --build
5 — Run Tests
# Offline unit + protocol tests (no credentials needed)
pytest -m "not sandbox"
# Live sandbox tests (requires valid credentials in .env)
pytest -m sandbox
Project layout (detail)
openfinance-mcp/
src/openfinance_mcp/
finicity_client.py
server.py
models.py
spec_defaults.py
config/settings.py
__main__.py
tests/
console/
app.py # FastAPI bridge (MCP client → browser)
static/ # index.html, console.js, console.css
run.sh / run.bat
docs/
console-screenshot.png
run.sh / run.bat
docker-compose.yml
pyproject.toml
Resolved spec decisions
| Question | Answer | Status |
|---|---|---|
| PSI path/body | POST /payments/customers/{id}/accounts/{aid}/payment-success-indicators, {"transaction":{"settleByDate","amount"}} |
confirmed |
| VOAI slug | voaHistory |
confirmed |
| TxPUSH body | {"callbackUrl": <url>} |
confirmed |
The console's Spec Decisions tab renders these live from the running server.
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.
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.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.