# BookMarket Research MCP Server Agent

An agentic Market Research Model Context Protocol (MCP) server for Digital, Audio, Print-on-Demand (POD), and Ebook publishing analytics.

Built with Python, FastMCP, Supabase, Airtable, Hugging Face, GitHub, DuckDuckGo Search, and Gemini API.

---

## 🛠️ Features

*   **DuckDuckGo Search integration** to perform live queries on market trends, competitor titles, and publisher dynamics.
*   **Gemini API processing** to clean, categorize, and synthesize raw search data into formatted intelligence reports.
*   **Supabase storage** to store reports persistently in a SQL table.
*   **Airtable Logging** to track campaigns and research jobs.
*   **Hugging Face Hub integration** to push research datasets or models directly.
*   **GitHub commits** to push final markdown reports straight to git repositories.

---

## ⚙️ Configuration

Copy `.env.example` to `.env` and fill in the required credentials:

```bash
cp .env.example .env
```

```ini
GEMINI_API_KEY=your_gemini_api_key
SUPABASE_URL=your_supabase_url
SUPABASE_KEY=your_supabase_key
AIRTABLE_PAT=your_airtable_personal_access_token
HF_TOKEN=your_hf_write_token
GITHUB_TOKEN=your_github_token
```

---

## 🚀 Usage

### 1. Installation

Install all required Python dependencies:

```bash
pip install -r requirements.txt
```

### 2. Run the MCP Server (stdio)

You can run the server directly using python:

```bash
python mcp_server.py
```

### 3. Registering with Gemini CLI

To load this server in the Gemini CLI, use the `gemini mcp add` command:

```bash
gemini mcp add bookmarket python /home/brettanthonysjoberg079/brettapps/bookmarket/agent/mcp_server.py
```