wealthapi-mcp
MCP server that enables AI assistants to answer portfolio questions about a wealthAPI account, with read-only tools for accounts, transactions, investments, dividends, and market data.
README
wealthapi-mcp
A small Model Context Protocol (MCP) server that lets Claude (or any MCP-compatible assistant) answer everyday portfolio questions about your wealthAPI account:
"What did I spend the most money on last month?" "How is my portfolio performing this year?" "Show me my five biggest positions." "Which of my investments lost the most value?"
It runs locally on your machine. Your bearer token never leaves your computer.
What is an MCP server?
MCP is an open protocol that lets an AI assistant call tools you control. The assistant
runs in one process (e.g. Claude Desktop); the MCP server runs as a separate process you
launch. They talk to each other over stdio using JSON-RPC: the assistant writes a
tools/call message to the server's stdin, the server runs the tool and writes the
result back to its stdout. That's the whole protocol.
Note the distinction: this is an MCP server, not an MCP plugin. There is no such thing as an MCP plugin — plugins (e.g. Claude Code plugins) extend the assistant itself, while an MCP server is a standalone process the assistant connects to over the protocol. Any MCP-compatible client (Claude Desktop, Claude Code, others) can use this server.
Concretely, this repo is a Node script. When Claude Desktop starts it, the script registers its tools, waits for tool calls, and when one arrives, makes an HTTPS request to the wealthAPI REST API on your behalf and returns a text summary. No state, no database, no server to maintain.
Read more at https://modelcontextprotocol.io.
The tools
Auth is a wealthAPI API token (wapi_key_…) carrying the read scopes (accounts:read,
investments:read, transactions:read, profile:read). All tools are read-only.
Accounts & cash flow
| Tool | Wraps | Answers |
|---|---|---|
list_accounts |
GET /api/v1/accounts |
"What accounts do I have?" |
get_account_balances |
GET /api/v1/accounts/balances |
"Show my cash balance history" |
get_portfolio_valuation |
GET /api/v1/accounts/valuation (+ historicValuations) |
"What is my portfolio worth (today / over time)?" |
list_transactions |
GET /api/v1/transactions |
"What did I spend most on last month?" |
detect_recurring_transactions |
GET /api/v1/transactions |
"What subscriptions / fixed costs do I have?" |
list_bookings |
GET /api/v1/bookings |
"Show my buys / sells / dividends" |
get_cash_flow_summary |
GET /api/v1/cashFlowAnalytics/history |
"Income vs spending per month" |
get_savings_rate |
GET /api/v1/cashFlowAnalytics/history |
"What's my savings rate?" |
Portfolio & performance
| Tool | Wraps | Answers |
|---|---|---|
list_investments |
GET /api/v2/investments |
"What do I own and what is it worth?" |
get_gainers_and_losers |
GET /api/v1/investments/gainersAndLosers |
"Best / worst performers" |
get_portfolio_performance |
POST /api/v2/performance |
"How did my portfolio do this year?" |
get_portfolio_allocation |
GET /api/v1/investments |
"How diversified am I by region / sector / asset type?" |
get_risk_metrics |
GET /api/v1/riskYieldMetrics (+ /investments) |
"How risky is my portfolio?" |
get_realized_gains |
GET /api/v1/performance/realizedGains |
"What did I realize this year?" (tax season) |
Dividends
| Tool | Wraps | Answers |
|---|---|---|
get_dividend_history |
GET /api/v1/dividends/history |
"How much dividend income did I get per year / month?" |
get_dividend_calendar |
GET /api/v1/dividends/calendar |
"When are my next dividend payments?" |
get_portfolio_yield |
GET /api/v1/dividends/portfolioYield |
"What's my dividend yield?" |
Market data & profile
| Tool | Wraps | Answers |
|---|---|---|
search_symbols |
GET /api/v2/symbols |
"Find Apple / this ISIN" |
get_quotes |
GET /api/v2/quotes |
"What's the current price?" |
get_security_fundamentals |
GET /api/v1/fundamentals/{isin} (+ /statistics) |
"Is this stock expensive?" (P/E, P/B, F-Score) |
whoami |
GET /api/v1/users/myself |
"Which account is this?" |
The server also registers prompts (monthly_review, savings_rate_check,
subscription_audit, fixed_costs_summary, financial_health_check) that chain these
tools into guided analyses.
Tool registration is gated by SUPPORTED_TOOLS/SUPPORTED_PROMPTS in src/index.ts —
remove a name there to disable it without deleting code.
Install
You need Node 20+ and pnpm.
git clone git@github.com:wealthAPI-eu/wealthapi-mcp.git
cd wealthapi-mcp
pnpm install
pnpm build
Get a personal bearer token from wealthAPI (your account → API settings).
Add the server to Claude Desktop's config
(~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"wealthapi": {
"command": "node",
"args": ["/absolute/path/to/wealthapi-mcp/build/index.js"],
"env": {
"BEARER_TOKEN": "paste-your-token-here"
}
}
}
}
Restart Claude Desktop. The active tools will appear in the tools panel of any chat, and you can ask portfolio questions in plain English.
How to add a new tool
The canonical example lives in src/tools/accounts.ts — list_accounts is ~50 lines
and shows the whole pattern.
- Create or open the relevant file under
src/tools/and write aregister*Tools(server, client)function that callsserver.registerTool(...). Each call takes a name, a{ description, inputSchema }object (Zod for the input schema), and an async handler that returns{ content: [{ type: "text", text }] }. - Import and call your
register*Toolsfromsrc/index.ts. pnpm buildand restart Claude Desktop.
That's it. No code generation, no registration files, no boilerplate.
Layout
src/
index.ts Bootstrap: McpServer + StdioServerTransport + register*Tools(...)
config.ts Reads BEARER_TOKEN from env; API base URL is fixed to production
api/
client.ts fetch wrapper: GET/POST, query strings, retry on 5xx/429
types.ts TS interfaces for API response shapes
tools/
accounts.ts list_accounts, get_account_balances, get_portfolio_valuation
allocation.ts get_portfolio_allocation
cash-flow.ts get_cash_flow_summary, get_savings_rate, detect_recurring_transactions
dividends.ts get_dividend_history, get_dividend_calendar, get_portfolio_yield
investments.ts list_investments, get_gainers_and_losers
performance.ts get_portfolio_performance, get_realized_gains
risk.ts get_risk_metrics
securities.ts search_symbols, get_quotes, get_security_fundamentals
transactions.ts list_transactions, list_bookings
user.ts whoami
util/
logger.ts Pino → stderr (stdout is reserved for MCP)
format.ts formatCurrency, formatDate, formatTable
test/
api/
client.test.ts Vitest unit test for the query-string builder
tools/
*.test.ts Unit tests for the pure aggregation helpers
Run tests
pnpm test
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.