Skip to main content
Glama
lynx-mi

lynx-mi/lynx-mi-mcp

by lynx-mi
README.md
# 🐆 Lynx MI — MCP Server

> **Model Context Protocol server for Lynx Market Intelligence**
>
> Give your AI coding assistant real-time access to SEC insider trading data, market sentiment, political intelligence, and more.

[![npm version](https://img.shields.io/npm/v/@lynx-market-intel/mcp-server)](https://www.npmjs.com/package/@lynx-market-intel/mcp-server)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

---

## What is this?

This MCP server connects AI coding assistants like **Claude**, **Cursor**, **Windsurf**, and **Cline** directly to [Lynx MI](https://lynx-mi.com) — a market intelligence platform that tracks SEC insider trades, congressional stock transactions, lobbying activity, and more.

Once connected, you can ask your AI assistant questions like:

- *"What insider trades happened at NVDA this week?"*
- *"Show me the top insiders by performance over 90 days"*
- *"Which sectors are insiders buying the most right now?"*
- *"Are any insiders trading near upcoming earnings dates?"*
- *"What's the Nexus political intelligence score for AAPL?"*

## Prerequisites

- **Node.js** 18+
- A **Lynx MI** account with an **Advanced subscription** or higher
- A **Personal API Key** (generate at [lynx-mi.com/settings](https://lynx-mi.com/settings))

## Quick Setup

### Claude Desktop

Add this to your `claude_desktop_config.json`:

**macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
**Windows:** `%APPDATA%\Claude\claude_desktop_config.json`

```json
{
  "mcpServers": {
    "lynx-mi": {
      "command": "npx",
      "args": ["-y", "lynx-mi-mcp-server"],
      "env": {
        "LYNX_MI_API_KEY": "sk_live_your_key_here"
      }
    }
  }
}
```

### Cursor

Go to **Settings → MCP Servers → Add new server** and use:

```json
{
  "command": "npx",
  "args": ["-y", "@lynx-mi/mcp-server"],
  "env": {
    "LYNX_MI_API_KEY": "sk_live_your_key_here"
  }
}
```

### Windsurf / Cline

Add to your MCP configuration:

```json
{
  "lynx-mi": {
    "command": "npx",
    "args": ["-y", "@lynx-mi/mcp-server"],
    "env": {
      "LYNX_MI_API_KEY": "sk_live_your_key_here"
    }
  }
}
```

### Google Gemini CLI

Add to your `~/.gemini/settings.json`:

```json
{
  "mcpServers": {
    "lynx-mi": {
      "command": "npx",
      "args": ["-y", "@lynx-mi/mcp-server"],
      "env": {
        "LYNX_MI_API_KEY": "sk_live_your_key_here"
      }
    }
  }
}
```

## Available Tools

### 📊 Trades
| Tool | Description |
|------|-------------|
| `get_latest_trades` | Recent SEC insider trades with filters |
| `get_trades_by_ticker` | All insider activity for a stock |
| `get_trades_by_insider` | All trades by a specific insider |
| `search_insider` | Search insiders by name |
| `search_ticker` | Search tickers by symbol or name |

### 📈 Market Intelligence
| Tool | Description |
|------|-------------|
| `get_market_sentiment` | Current market insider buy/sell ratio |
| `get_top_movers` | Stocks with highest insider volume |
| `get_insider_profile` | Detailed insider profile & stats |

### 🔬 Pro Analytics
| Tool | Description |
|------|-------------|
| `get_sector_flow` | Money flow by sector (leading indicator) |
| `get_correlations` | Hidden company connections via shared insiders |
| `get_insider_network` | Network graph of insider-company relationships |
| `get_network_top` | Most connected insiders & companies |
| `get_trade_heatmap` | Volume heatmap by ticker |
| `get_earnings_proximity` | Trades near upcoming earnings (⚠️ high signal) |
| `get_filing_diff` | Compare insider's last two SEC filings |
| `get_conviction_score` | Trade conviction score breakdown |

### 🏆 Leaderboard
| Tool | Description |
|------|-------------|
| `get_leaderboard` | Top insiders by 30/90-day returns |
| `get_insider_rank` | Specific insider's rank & stats |

### 🏛️ Nexus (Political Intelligence)
| Tool | Description |
|------|-------------|
| `get_nexus_score` | Combined political + lobbying + insider score |
| `get_congressional_trades` | US Congress member stock trades |
| `get_lobbying_activity` | Corporate lobbying reports & spend |

## Time-Series & Supply Chain Intelligence

In addition to real-time insider trading, Lynx MI provides deep, historical time-series data and relational graphs to power advanced AI quantitative analysis:

- **Congressional & Executive Trading (STOCK Act)**: Query historical time-series data of stock transactions by US Senators, Representatives, and their families. Track portfolio changes aligned with specific legislative events or committee assignments to uncover political information advantages.
- **Corporate Lobbying Expenditures (LDA)**: Access quarter-by-quarter historical time-series of lobbying spend. Track which bills and government agencies specific companies are targeting, offering predictive insights into upcoming regulatory headwinds or tailwinds.
- **Supplier & Customer Supply Chain Graph**: Extract complex B2B dependencies. Map out tier-1 and tier-2 suppliers, key customer relationships, and revenue exposure. If one company inside the network crashes or secures a massive contract, your AI assistant can trace the direct financial impact down the supply chain using our Nexus endpoints.

## Example Conversations

### Research a stock
> **You:** What's the insider trading activity at Tesla?
>
> **Claude:** *Uses `get_trades_by_ticker` with ticker="TSLA"*
> I found 12 insider trades at Tesla in the last 90 days...

### Detect signals
> **You:** Are there any suspicious trades near earnings?
>
> **Claude:** *Uses `get_earnings_proximity`*
> I found 5 trades within 7 days of upcoming earnings...

### Political intelligence
> **You:** What's the political risk for Apple?
>
> **Claude:** *Uses `get_nexus_score` with ticker="AAPL"*
> Apple's Nexus score is 72/100, driven by heavy lobbying spend...

## Getting Your API Key

1. Go to [lynx-mi.com](https://lynx-mi.com) and create an account
2. Subscribe to the **Advanced** plan or higher
3. Navigate to **Settings → API Key**
4. Click **Generate API Key**
5. Copy the `sk_live_...` key into your MCP configuration

## Development

```bash
# Install dependencies
npm install

# Run in development mode
LYNX_MI_API_KEY=sk_live_... npm run dev

# Build for production
npm run build

# Test the built server
LYNX_MI_API_KEY=sk_live_... npm start
```

## Pricing

The MCP server is **free and open source**. You only pay for the Lynx MI subscription that provides the API key:

| Plan | Price | MCP Access |
|------|-------|------------|
| Free | $0/mo | ❌ |
| Recreational | $9.99/mo | ❌ |
| **Advanced** | **$24.99/mo** | **✅ All tools** |
| Enterprise | $49.99/mo | ✅ All tools + Nexus |

## License

MIT © [Lynx MI](https://lynx-mi.com)

TDQS

A3.6/5.0

Scored across 21 tools

Disambiguation5/5

Every tool has a clearly distinct purpose with no ambiguity, as each targets a specific aspect of the insider trading and political intelligence domain (e.g., get_congressional_trades for congressional activity, get_conviction_score for trade analysis, get_orrelations for network insights). The descriptions precisely differentiate their functions, ensuring agents can easily select the right tool without confusion.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern using snake_case, starting with 'get_' or 'search_' followed by a descriptive noun phrase (e.g., get_insider_profile, search_ticker). This uniformity makes the set predictable and readable, aiding agents in understanding and navigating the tools efficiently.

Tool Count4/5

With 21 tools, the count is slightly high but reasonable for the comprehensive scope of insider trading and political intelligence analysis. Each tool appears to serve a specific, non-redundant function, though it may feel heavy for simpler use cases; overall, it's well-scoped for the domain's complexity.

Completeness5/5

The tool surface provides complete coverage of the domain, including data retrieval (e.g., trades, profiles, scores), analysis (e.g., correlations, heatmaps), and search capabilities (insiders, tickers). It supports full workflows from discovery to deep analysis with no obvious gaps, enabling agents to handle a wide range of tasks without dead ends.

Maintenance

ActivityInactive
ResponsivenessNo issues