FinImpulse MCP Server

A Model Context Protocol (MCP) server that provides structured financial market data to AI systems and development tools through the FinImpulse API.

It allows AI systems to access FinImpulse data — including stocks, ETFs, mutual funds, fundamentals, and market indicators — using a standardized MCP interface. The server acts as a bridge between the FinImpulse API and any MCP-compatible client, enabling LLMs and automation tools to retrieve financial data in a consistent format without building custom integrations.

Hosted server available. You can connect any Streamable HTTP MCP client directly to https://mcp.finimpulse.com/mcp and authenticate via OAuth — no API token setup required. See Quick Start below.

How It Works

AI Client (Claude / ChatGPT / Cursor / ...)
        ↓
   MCP Server
        ↓
  FinImpulse API
        ↓
  Financial Data

The MCP server translates tool calls from AI clients into FinImpulse API requests and returns structured responses that AI models can interpret. This architecture allows AI tools to access financial datasets without implementing direct API integrations.

Key Features

• Structured financial market data via MCP
• Global coverage of stocks, ETFs, and mutual funds
• Standardized tools for AI agents and developer environments
• Compatible with major MCP clients and AI platforms

Use Cases

• AI investment assistants
• Automated financial research workflows
• AI-assisted financial dashboards
• Trading assistants
• Portfolio analysis tools
• Developer environments retrieving market data

Example prompt — once connected, an AI system can retrieve financial data using natural language:

Compare Microsoft and Google in terms of revenue growth and profitability.

Requirements

You can use FinImpulse MCP in two ways:

• Hosted (recommended): no requirements other than a FinImpulse account. Authentication is handled via OAuth in the browser — no manual API token management.
• Local (stdio): Node.js 18+ and a FinImpulse API token (obtain it in the FinImpulse Dashboard).

Quick Start

There are two ways to connect: use the hosted server at https://mcp.finimpulse.com/mcp (Streamable HTTP + OAuth), or run the server locally via npx.

---

Option 1 — Hosted server (OAuth)

The FinImpulse-hosted MCP endpoint is available at:

https://mcp.finimpulse.com/mcp

It uses the Streamable HTTP transport and OAuth 2.0 — when you first connect, your MCP client will open a browser window to authorize access against your FinImpulse account. The client then stores the access token and forwards it automatically on every request. No API token needs to be configured by hand.

Cursor

Open ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-level) and add:

{
  "mcpServers": {
    "finimpulse": {
      "url": "https://mcp.finimpulse.com/mcp"
    }
  }
}

VS Code (Copilot)

Open your VS Code settings.json and add:

{
  "mcp": {
    "servers": {
      "finimpulse": {
        "url": "https://mcp.finimpulse.com/mcp"
      }
    }
  }
}

Claude Desktop

Claude Desktop does not yet speak Streamable HTTP natively. Bridge through mcp-remote, which handles the OAuth flow in your browser and forwards requests over stdio:

{
  "mcpServers": {
    "finimpulse": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.finimpulse.com/mcp"]
    }
  }
}

ChatGPT

In the ChatGPT desktop app go to Settings → Connectors → Add custom connector and enter:

• URL: https://mcp.finimpulse.com/mcp
• Authentication: OAuth

Windsurf / Cline / other Streamable-HTTP clients

Any MCP client that supports the Streamable HTTP transport can use the URL directly:

https://mcp.finimpulse.com/mcp

For clients that only support stdio, use the mcp-remote bridge shown in the Claude Desktop example.

---

Option 2 — Local (stdio via npx)

If you prefer to run the server in-process on your machine, use the npm package. You'll need a FinImpulse API token from the FinImpulse Dashboard.

Claude Desktop

Open claude_desktop_config.json:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

Add the server configuration:

{
  "mcpServers": {
    "finimpulse": {
      "command": "npx",
      "args": ["-y", "finimpulse-mcp-server"],
      "env": {
        "API_TOKEN": "your-api-token"
      }
    }
  }
}

Claude Code

claude mcp add finimpulse -- npx -y finimpulse-mcp-server

Then set the API_TOKEN environment variable or pass it inline.

Cursor

Open ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-level) and add:

{
  "mcpServers": {
    "finimpulse": {
      "command": "npx",
      "args": ["-y", "finimpulse-mcp-server"],
      "env": {
        "API_TOKEN": "your-api-token"
      }
    }
  }
}

Windsurf

Open ~/.codeium/windsurf/mcp_config.json and add:

{
  "mcpServers": {
    "finimpulse": {
      "command": "npx",
      "args": ["-y", "finimpulse-mcp-server"],
      "env": {
        "API_TOKEN": "your-api-token"
      }
    }
  }
}

VS Code (Copilot)

Open your VS Code settings.json and add:

{
  "mcp": {
    "servers": {
      "finimpulse": {
        "command": "npx",
        "args": ["-y", "finimpulse-mcp-server"],
        "env": {
          "API_TOKEN": "your-api-token"
        }
      }
    }
  }
}

Gemini CLI

gemini mcp add finimpulse -- npx -y finimpulse-mcp-server

Cline

Open Cline settings, go to MCP Servers, and add a new server with this config:

{
  "mcpServers": {
    "finimpulse": {
      "command": "npx",
      "args": ["-y", "finimpulse-mcp-server"],
      "env": {
        "API_TOKEN": "your-api-token"
      }
    }
  }
}

Any MCP-Compatible Client (stdio)

API_TOKEN=your-api-token npx finimpulse-mcp-server

Available Tools

Search & Discovery

Tool	Description
get_search	Cross-asset search for stocks, ETFs, and mutual funds. Powers ticker autocomplete and discovery flows.

Market Data

Tool	Description
get_summary	Comprehensive single-call snapshot: identity, market data, fundamentals, ownership, analyst coverage, and fund metrics.
get_histories	Historical time-series: prices, dividends, splits, and capital gains with configurable date range and interval.
get_profile	Asset profile and descriptive metadata: identity, classification, governance, and fund-specific attributes.
get_news	Latest news and press releases with headlines, descriptions, related tickers, and provider info.

Statistics & Performance

Tool	Description
get_statistics_general	Key statistics and performance indicators: valuation, dividends, risk, trailing returns.
get_statistics_risks	Risk-adjusted performance metrics (Alpha, Beta, Sharpe, Treynor, R-squared) across 3y/5y/10y horizons.
get_statistics_annual_returns	Historical annual total returns by calendar year with quarterly breakdowns.

Financial Statements

Tool	Description
get_financials_general	Unified financial statements and valuation measures across reporting intervals.
get_financials_income_statement	Detailed Income Statement: revenue, expenses, net income (annual/quarterly/trailing).
get_financials_balance_sheet	Balance Sheet: financial position and capital structure snapshot.
get_financials_cash_flow	Cash Flow Statement: operating, investing, and financing cash flows.
get_financials_valuation_measures	Valuation Measures: P/E, EV/EBITDA, P/B, and related metrics.

Holders & Ownership

Tool	Description
get_holders_general	High-level ownership snapshot: institutional and insider holdings percentages.
get_holders_institutional	Paginated list of institutional holders with position size, percent held, and value.
get_holders_mutual_funds	Mutual funds and ETFs that hold a given stock.
get_holders_insiders	Insider transaction records: actions, roles, and reported positions.
get_holders_insiders_transactions	Detailed insider transactions with share counts, values, and filer identity.

Fund Holdings

Tool	Description
get_holdings_general	Portfolio composition and holdings-level aggregates: asset allocation, sector exposure.
get_holdings_top_holdings	Top holdings list with ticker, name, and weight as % of total net assets.

Analyst Coverage & Ratings

Tool	Description
get_analysis_analysts	Analyst coverage records per analyst: scores, ratings, sentiment, price targets, and announcement dates.
get_analysis_earnings	Earnings and estimate data: EPS actuals, earnings vs. revenue, EPS trends, revisions, and growth metrics.
get_analysis_recommendations	Analyst recommendation breakdown over time: Strong Buy, Buy, Hold, Sell, and Strong Sell counts per period.
get_analysis_upgrades_downgrades	Analyst rating action feed: upgrades, downgrades, reiterations, and price target changes with from/to grades.

Options

Tool	Description
get_options_expirations	Available option expiration dates for an underlying symbol.
get_options_chain	Full option chain for a given expiration: calls and puts with pricing, bid/ask, volume, open interest, and implied volatility.
get_options_contracts	Single option contract snapshot: pricing, bid/ask, volume, open interest, implied volatility, moneyness, and Greeks.

Configuration

When using the hosted server (https://mcp.finimpulse.com/mcp) no environment variables are required — your MCP client handles OAuth automatically.

The variables below apply only when running the server locally (stdio via npx or your own deployment):

Environment Variable	Required	Description
API_TOKEN	Yes (local stdio)	Your FinImpulse API token

Development

git clone https://github.com/finimpulse/mcp-server-typescript.git
cd mcp-server-typescript
npm install
npm run build

Run locally:

API_TOKEN=your-api-token npm run cli

Security

Keep your API token secure. Never expose API credentials in public repositories. Always store them using environment variables.

Disclaimer

The AI-powered features and examples provided through FinImpulse MCP are intended for informational and research purposes only. Any insights, summaries, or suggestions generated by AI systems using FinImpulse data should be considered analytical tools, not financial, investment, or trading advice.

FinImpulse does not provide investment recommendations and does not guarantee the accuracy, completeness, or suitability of AI-generated outputs. Users are solely responsible for evaluating any information produced by AI agents or applications built on top of FinImpulse MCP. Financial decisions should always be made based on independent analysis or consultation with qualified financial professionals.

Learn More

• FinImpulse API Documentation
• Model Context Protocol

License

ISC