Product Hunt Analyst

Independent project: pulls data from Product Hunt, analyzes hot product trends from the past 1 / 2 / 3 / 4 weeks or 1 month, and provides a visual web interface.

It can be integrated with tools such as Codex, Cursor, Claude Desktop, and Workbuddy via MCP / CLI / HTTP API.

Features

• Pull Product Hunt products by time window (sorted by upvotes)

• Automatically generate trend reports: Top products, category distribution, daily launch volume, topic keywords, and textual insights

• React visualization dashboard (line charts, bar charts, leaderboard)

• MCP server for direct invocation by AI agents

• CLI script that outputs JSON / Markdown

Related MCP server: GitHub Trending Service

Quick Start

1. Install Dependencies

cd product-hunt-analyst
python -m venv .venv
# Windows
.venv\Scripts\activate
# macOS/Linux
source .venv/bin/activate

pip install -r requirements.txt

2. Configure the Product Hunt Token

1. Create an application at Product Hunt OAuth Apps

2. Copy the Developer Token

3. Copy .env.example to .env and fill it in:

PRODUCT_HUNT_TOKEN=your_token_here

When no Token is configured, the Web UI automatically switches to Demo mode and displays sample data.

3. Command Line Analysis

# JSON 报告
python scripts/analyze.py --period 2w

# Markdown 报告
python scripts/analyze.py --period 1m --markdown

# 仅拉取原始数据
python scripts/fetch_posts.py --period 3w --compact

4. Launch the Web Visualization

Method A: Development mode (frontend and backend separated)

# 终端 1 - API
python -m backend.app

# 终端 2 - 前端
cd frontend
npm install
npm run dev
# 打开 http://localhost:5173

Method B: Production mode (single port)

cd frontend && npm install && npm run build
cd ..
python -m backend.app
# 打开 http://127.0.0.1:8765

Time Range

MCP Integration (Codex / Cursor / Claude)

Codex

pip install -r requirements.txt
export PRODUCT_HUNT_TOKEN=your_token
codex mcp add productHuntAnalyst --command python --arg mcp/server.py

After linking or copying skill/SKILL.md to the Codex skills directory, you can trigger it using $product-hunt-analyst.

Cursor

Add the following to .cursor/mcp.json in the project root:

{
  "mcpServers": {
    "productHuntAnalyst": {
      "command": "python",
      "args": ["mcp/server.py"],
      "cwd": "E:/codex/product-hunt-analyst",
      "env": {
        "PRODUCT_HUNT_TOKEN": "your_token"
      }
    }
  }
}

Claude Desktop

Register the same MCP server configuration in claude_desktop_config.json.

MCP Tools

HTTP API

Workbuddy / Automation

Just call it via HTTP or CLI:

curl "http://127.0.0.1:8765/api/analyze?period=4w"
python scripts/analyze.py -p 4w --markdown > report.md

Project Structure

product-hunt-analyst/
├── backend/           # FastAPI + PH API 客户端 + 分析引擎
├── frontend/          # React + Vite + Recharts 可视化
├── mcp/               # MCP Server
├── scripts/           # CLI 工具与测试
├── skill/             # Codex/Cursor Agent Skill
├── agents/            # OpenAI Agents 配置
└── requirements.txt

Testing

python scripts/test_product_hunt_analyst.py

Notes

• The Product Hunt API is rate limited (around 6250 complexity points / 15 minutes)

• The data is for research and reference only and does not constitute investment or business decision advice

• Please comply with the Product Hunt API Terms

License

MIT