FinLynq
open-source personal finance app with a first-party MCP server. 91 HTTP tools (OAuth 2.1 + DCR) and 87 stdio tools cover transactions, budgets, accounts, portfolio analytics, FX conversion, loans, subscriptions, goals, importers, and rules. Users self-host with Docker + PostgreSQL or use the managed cloud
README
Finlynq — Personal Finance App
An open-source personal finance web app with a first-party MCP server. Track income, expenses, budgets, investments, loans, and goals — then query your financial data from Claude, Cursor, Windsurf, or any MCP-compatible AI assistant.
"Track your money here, analyze it anywhere."
License: AGPL v3 · Repo: github.com/finlynq/finlynq · Support: GitHub Sponsors · Ko-fi
Try the Demo
No signup needed — a public demo account is always available on production:
- URL: finlynq.com
- Username:
demo(or emaildemo@finlynq.com— login accepts either) - Password:
finlynq-demo
Comes preloaded with 6 months of sample transactions, accounts, budgets, investments, and goals. Data resets nightly at 03:00 UTC, so feel free to change anything.
Connect it to Claude: in Claude → Customize → Connectors → "+" and paste https://finlynq.com/mcp. OAuth handles the rest — no config file.
Running Locally (Windows)
Prerequisites
- Node.js 18+
- PostgreSQL 14+ installed and running (the Windows service
postgresql-x64-*should start automatically) - Git
1. Clone & install
git clone https://github.com/finlynq/finlynq.git
cd finlynq/pf-app
npm install
2. Create your .env file
Copy .env.example to .env and fill in the values:
cp .env.example .env
Minimum required values for local dev:
NODE_ENV=development
APP_URL=http://localhost:3000
DATABASE_URL=postgres://<user>:<password>@localhost:5432/finlynq
PF_JWT_SECRET=<run: openssl rand -hex 32>
Windows tip: If you know a working PostgreSQL user from another project (e.g.
epm:epm_dev_password), use those credentials — you just need a user withCREATEDBprivilege.
3. Create the database
Open PowerShell in the pf-app directory and run:
# Set your psql path (adjust version number as needed)
$psql = "C:\Program Files\PostgreSQL\16\bin\psql.exe"
$env:PGPASSWORD = "<your-db-password>"
# Create the finlynq database (connect to an existing DB first, e.g. epm_agent_v3 or postgres)
"CREATE DATABASE finlynq;" | & $psql -U <your-user> -h 127.0.0.1 -p 5432 -d <existing-db>
Or open pgAdmin 4 and run:
CREATE DATABASE finlynq;
4. Push the schema
powershell -ExecutionPolicy Bypass -Command "npm run db:push:pg"
Note: On Windows, running
npmdirectly in PowerShell may fail with a script execution policy error. Always prefix withpowershell -ExecutionPolicy Bypass -Command "..."or run from Git Bash / Command Prompt.
5. Start the dev server
powershell -ExecutionPolicy Bypass -Command "npm run dev"
App will be available at http://localhost:3000
6. Create your account
Navigate to http://localhost:3000/cloud and click Create Account. Pick a username (3–254 chars, lowercase letters / digits / . @ + _ -) — email is optional.
Heads up: Finlynq encrypts everything with your password. There's no recovery key — if you forget your password, all data is wiped on reset. Adding an email lets you trigger a reset; without one you have no way to regain access.
Running Locally (macOS / Linux)
Prerequisites
- Node.js 18+
- PostgreSQL 14+ (
brew install postgresql@16on macOS, orapt install postgresqlon Ubuntu)
Steps
# Clone
git clone https://github.com/finlynq/finlynq.git
cd finlynq/pf-app
# Install dependencies
npm install
# Configure environment
cp .env.example .env
# Edit .env — set DATABASE_URL and PF_JWT_SECRET
# Create the database (assuming default postgres superuser)
psql -U postgres -c "CREATE DATABASE finlynq;"
# Push schema
npm run db:push:pg
# Start dev server
npm run dev
App will be available at http://localhost:3000
Docker (Self-hosted)
The easiest way to run Finlynq with zero manual setup:
git clone https://github.com/finlynq/finlynq.git
cd finlynq/pf-app
# Copy and configure env
cp .env.example .env
# Edit .env — at minimum set PF_JWT_SECRET
# Start app + PostgreSQL
docker compose up -d
App will be available at http://localhost:3000
The Docker Compose file starts both the app and a PostgreSQL 16 container. Data is persisted in a named volume (postgres_data).
Environment Variables
| Variable | Required | Description |
|---|---|---|
DATABASE_URL |
✅ | PostgreSQL connection string — activates managed mode |
PF_JWT_SECRET |
✅ | 64-char hex secret for JWT signing (openssl rand -hex 32) |
NODE_ENV |
development or production |
|
APP_URL |
Public URL of the app (used in email links) | |
PF_SMTP_HOST |
SMTP host for password reset emails (leave blank to disable) | |
NEXT_PUBLIC_GITHUB_REPO |
Link to your fork (shown in donation banner) |
Tech Stack
| Layer | Technology |
|---|---|
| Framework | Next.js 16 (App Router, Turbopack) + TypeScript |
| Database | PostgreSQL 16 + Drizzle ORM (22 tables) |
| UI | Tailwind CSS + shadcn/ui v4 (@base-ui/react) |
| Charts | Recharts + custom Sankey SVG |
| Animations | Framer Motion |
| Theming | next-themes (dark/light/system) |
| MCP | @modelcontextprotocol/sdk (86 tools, Streamable HTTP + stdio) |
| Auth | OAuth 2.1 + DCR (MCP), session-cookie + Bearer API key (REST) |
| Prices | Yahoo Finance API + CoinGecko (crypto) — both free, no key needed |
MCP Server
Finlynq includes a first-party MCP server with 86 tools for querying and managing your financial data from AI assistants — accounts, transactions, budgets, goals, loans, portfolio, subscriptions, FX rates, rules, splits, bulk edits, and CSV/OFX file imports.
Connect via Claude Desktop (stdio)
Add to ~/.claude/claude_desktop_config.json:
{
"mcpServers": {
"finlynq": {
"command": "npx",
"args": ["tsx", "/path/to/pf-app/mcp-server/index.ts"],
"env": {
"DATABASE_URL": "postgres://epm:epm_dev_password@localhost:5432/finlynq"
}
}
}
}
Connect via Claude Web / Mobile (HTTP)
- Go to Settings > API Keys in the app and generate a key (prefix
pf_) - Add the MCP server URL in your AI assistant:
https://your-domain.com/api/mcp - Use Bearer token auth with your generated key
The .well-known/mcp.json server card at the root enables auto-discovery.
Import Your Data
Go to /import in the app. Supported formats:
| Format | Notes |
|---|---|
| CSV | Preview with column mapping + deduplication |
| Excel (.xlsx/.xls) | Visual column mapper |
| Bank statement table extraction | |
| OFX / QFX | fitId-based deduplication |
All imports include a deduplication engine that fingerprints transactions to prevent duplicates.
Pages
| Page | Path | Description |
|---|---|---|
| Dashboard | /dashboard |
Net worth, health score, sparklines, weekly recap, spotlight alerts |
| Transactions | /transactions |
Full list with filters, bulk edit, splits |
| Budgets | /budgets |
Monthly budgets with envelope-style rollover |
| Goals | /goals |
Financial goals with progress tracking |
| Accounts | /accounts |
All accounts grouped by type |
| Portfolio | /portfolio |
Holdings, allocation, benchmarking, crypto |
| Loans | /loans |
Amortization schedules, payoff scenarios |
| Reports | /reports |
Income statement, balance sheet, Sankey, YoY |
| Tax | /tax |
TFSA/RRSP/RESP room + RRSP vs TFSA calculator |
| Chat | /chat |
Natural language financial queries with inline charts |
| Subscriptions | /subscriptions |
Recurring subscription tracker with auto-detection |
| Calendar | /calendar |
Monthly bill calendar |
| Scenarios | /scenarios |
What-if modeling (home, savings, debt, income) |
| FIRE | /fire |
FIRE/retirement calculator + Monte Carlo simulation |
| Import | /import |
Multi-format import |
| MCP Guide | /mcp-guide |
Setup guide (Claude Desktop / Web / Mobile / API) |
| API Docs | /api-docs |
Developer API documentation |
| Settings | /settings |
Categories, API keys, backup/restore |
Scripts
| Script | Description |
|---|---|
npm run dev |
Start dev server at http://localhost:3000 |
npm run build |
Production build |
npm run build:mcp |
Build MCP server |
npm run db:push:pg |
Push schema to PostgreSQL |
npm run db:generate:pg |
Generate Drizzle PG migrations |
Project Structure
pf-app/
src/
app/ # 21 pages + 50+ API routes
components/ # Nav, unlock gate, setup wizard, sparkline, sankey, shadcn/ui
db/ # Schema (22 tables), PostgreSQL adapter, Drizzle ORM
lib/ # Business logic (parsers, calculators, AI chat engine, MCP tools)
mcp-server/ # MCP server v3 (86 tools, stdio + Streamable HTTP)
drizzle-pg/ # PostgreSQL migration files
mobile/ # React Native (Expo) mobile app
docs/ # Getting started, FAQ, mobile setup guide
Contributing
Contributions are welcome! Please read CONTRIBUTING.md first.
- Bug reports → GitHub Issues
- Feature requests → GitHub Discussions
- Pull requests → fork, branch, PR against
main
Support the Project
Finlynq is free and open-source (AGPL v3). If it's useful to you, consider supporting development:
- ⭐ Star the repo
- 💛 GitHub Sponsors
- ☕ Ko-fi
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.
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.
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.
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.