Bolagsinfo × Notion MCP Server
Connects Bolagsinfo.com's Nordic B2B lead intelligence platform to Notion, enabling AI agents to search, push, and enrich company leads directly into Notion databases for pipeline management.
README
🏢 Bolagsinfo × Notion MCP Server
DEV.to Notion MCP Challenge 2026
An MCP (Model Context Protocol) server that connects Bolagsinfo.com — a Nordic B2B lead intelligence platform with 2.4 million+ companies — to Notion. This lets AI agents like Claude, Windsurf, and Cursor search Nordic company leads, push them to Notion databases, enrich them with AI-powered Buy Signals (Köpsignaler), and manage a complete sales pipeline — all through natural language conversation. No clicking, no copy-pasting, no tab-switching. Just tell your AI assistant what you need and watch your Notion CRM fill up with qualified leads.
⚡ 5-Minute Setup
1. Clone & install
git clone https://github.com/cedendahlkim/bolagsinfo-notion-mcp.git
cd bolagsinfo-notion-mcp
pip install -r requirements.txt
2. Get your Notion API key
- Go to notion.so/my-integrations
- Click "New integration"
- Name it
Bolagsinfo MCP, select your workspace - Copy the Internal Integration Secret (starts with
ntn_) - Important: Share your target Notion page with this integration (click ••• → "Connect to" → "Bolagsinfo MCP")
3. Configure environment
cp .env.example .env
# Edit .env and add your NOTION_API_KEY
4. Run the server
# Stdio transport (for Claude Desktop / Windsurf / Cursor)
python server.py
# SSE transport (for Docker / remote)
python server.py --sse
# Or with Docker
docker-compose up
5. Connect to your AI assistant
Claude Desktop — add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"bolagsinfo": {
"command": "python",
"args": ["server.py"],
"cwd": "/path/to/bolagsinfo-notion-mcp",
"env": {
"NOTION_API_KEY": "ntn_your_key_here"
}
}
}
}
Windsurf — add to MCP server settings:
{
"mcpServers": {
"bolagsinfo": {
"command": "python",
"args": ["/path/to/bolagsinfo-notion-mcp/server.py"],
"env": {
"NOTION_API_KEY": "ntn_your_key_here"
}
}
}
}
🛠️ MCP Tools
| Tool | Description |
|---|---|
search_leads |
Search 2.4M+ Nordic companies by name, city, industry, Buy Signal Score |
create_lead_database |
Create a Notion database with the complete Bolagsinfo lead schema |
push_leads_to_notion |
Push companies (by org number) into Notion with smart upsert |
enrich_notion_lead |
Refresh a Notion lead with latest data from Bolagsinfo |
get_pipeline_summary |
Get pipeline analytics: status breakdown, scores, email coverage |
💬 Example Conversation
Here's a real workflow — from search to pipeline analytics — in one conversation:
Step 1: Search for leads
You: Hitta 10 e-handelsföretag i Stockholm med hög köpsignal
AI: Found 10 e-commerce companies in Stockholm:
Company Buy Signal Industry Nordic E-com AB 87 info@nordicecom.se E-handel ShopFlow Sweden 82 hello@shopflow.se Digital handel ... ... ... ...
Step 2: Create a Notion database
You: Skapa en Notion-databas för dessa leads på min sida
AI: ✅ Created "Bolagsinfo Leads" database. Database ID:
abc123...
Step 3: Push leads to Notion
You: Pusha alla 10 företagen till databasen
AI: ✅ Synced 10 leads to Notion:
- Created: 10
- Updated: 0
- Errors: 0
Step 4: Check pipeline
You: Visa pipeline-sammanfattning
AI: 📊 Pipeline Summary:
- Total leads: 10
- Status: New (10)
- Avg Buy Signal Score: 78.5
- Email coverage: 80%
- Top municipality: Stockholm (10)
Step 5: Enrich a specific lead
You: Berika det första företaget med färsk data
AI: ✅ Enriched "Nordic E-com AB":
- Buy Signal Score: 87
- Digital Maturity: 72
- CMS: WooCommerce
- Has Chatbot: No ← opportunity!
📸 Notion Database Result
After pushing leads, your Notion database looks like this:

Each lead includes company info, Buy Signal Score, AI Category, Digital Maturity, tech stack analysis, and pipeline status — all synced from Bolagsinfo.com.
🏗️ Architecture
┌──────────────┐ MCP (stdio/SSE) ┌──────────────────────┐
│ AI Agent │ ◄──────────────────────► │ Bolagsinfo MCP │
│ (Claude, │ │ Server (FastMCP) │
│ Windsurf, │ │ │
│ Cursor) │ │ ┌────────────────┐ │
└──────────────┘ │ │ 5 MCP Tools │ │
│ └───────┬────────┘ │
│ │ │
│ ┌───────▼────────┐ │
│ │ Services │ │
│ │ • Bolagsinfo │──────► bolagsinfo.com API
│ │ • Notion │──────► Notion API
│ └────────────────┘ │
└──────────────────────┘
🔑 Key Features
- 2.4M+ companies — Search the largest Nordic B2B database
- AI Buy Signals — Every company scored 0-100 on purchase likelihood
- Smart upsert — No duplicates. Updates existing leads, creates new ones.
- Rate limit aware — Respects Notion's 3 req/s limit with async semaphore
- Full CRM schema — 18 fields including tech stack, digital maturity, chatbot detection
- Pipeline analytics — Status breakdown, score averages, email coverage in seconds
- Zero config — Just add your Notion API key and go
📊 What is a Buy Signal Score?
Bolagsinfo.com analyzes every company across multiple dimensions:
- Digital presence — Website quality, SSL, mobile optimization
- Tech stack gaps — Missing chatbot, outdated CMS, no analytics
- Growth signals — Employee growth, new registrations, industry trends
- Engagement readiness — Has email, social presence, decision-maker accessibility
A score of 70+ means the company is a hot lead — they have clear digital gaps that your services can fill.
🇸🇪 Built by Gracestack AB
Gracestack AB builds AI-powered tools for B2B sales in the Nordics. Bolagsinfo.com is our flagship product — used by sales teams across Sweden and Norway to find, qualify, and close deals faster.
Kim Cedendahl — Founder & Developer
📄 License
MIT — use it, fork it, build on it.
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。