Fourth Brain Demo
An MCP server that connects Claude.ai to a Notion-based marketing knowledge base, enabling search and retrieval across specialized domains like enterprise platforms and competitive positioning. It provides tools for RAG-style Q\&A and content browsing to assist with drafting RFPs and value propositions.
README
Fourth Brain Demo
Notion-backed MCP server for Fourth's marketing knowledge base. Connects Claude.ai to curated marketing content via the Model Context Protocol.
Quick Start
1. Set up Notion
- Create a Notion Internal Integration at https://www.notion.so/my-integrations
- Create a parent page and share it with the integration
- Copy
.env.exampleto.env, add your token and parent page ID - Populate the knowledge base:
pip install -r requirements.txt
python populate_notion.py --parent-page <page_id>
- Copy the database IDs from the output into
.env
2. Run locally
python server.py
Server starts at http://localhost:8000/mcp
3. Test with MCP Inspector
fastmcp dev server.py
4. Deploy to Railway
# Push to GitHub
git remote add origin https://github.com/Rev4nchist/fourth-brain-demo.git
git push -u origin main
# Railway auto-deploys from GitHub
# Set env vars in Railway dashboard
5. Connect Claude.ai
- Go to Claude.ai > Settings > Integrations
- Add Custom Connector:
https://your-app.up.railway.app/mcp - Create a Project, upload skills files from
skills/
Architecture
Claude.ai --> MCP (Streamable HTTP) --> Notion API
|
FastMCP Server
- 5 tools (search, browse, get, ask, list)
- 3 prompts (product_qa, rfp_draft, meeting_prep)
- TTL cache (300s)
Tools
| Tool | Description |
|---|---|
search_knowledge |
Search across all 6 knowledge databases |
browse_library |
Browse knowledge structure by area |
get_document |
Retrieve full page content |
ask_question |
RAG-style Q&A with confidence scoring |
list_content_areas |
List all knowledge domains |
Knowledge Domains
| Domain | Content |
|---|---|
| Enterprise Platform | Platform overview, 5 pillars, modules, architecture |
| Value Propositions | Per-persona value props (CFO/COO/CHRO/CTO/CEO) |
| Integration Guide | 200+ integrations by category |
| Competitive Positioning | Battle cards, win/loss analysis |
| RFP Responses | Pre-drafted RFP answers |
| FAQ & Product Q&A | Common product questions |
Environment Variables
See .env.example for all required variables.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。