Nemo
An MCP server that enables users to transform AI conversations into a structured, searchable knowledge base by saving ideas, code snippets, bookmarks, and reminders. It supports persistent storage through Supabase or PostgreSQL and includes webhook integrations for automating workflows with external tools.
README
🧠 Nemo — MCP Server
Turn your AI conversations into a structured, searchable knowledge base.
An MCP (Model Context Protocol) server that lets you save conversations, ideas, code snippets, bookmarks, and reminders directly from Claude, ChatGPT, or any MCP-compatible AI assistant.
📁 Repository Layout
src/— Runtime server codedb/— Public database schema and migrationsdocs/— Product-facing public docsnotes/— Public work-in-progress notes, mockups, and experiments
✨ What It Does
You're chatting with Claude on your phone and the conversation is brilliant. Instead of losing it, you say:
"Save this in Nemo under DevOps, tag it docker and kubernetes"
Claude calls your MCP server, and your knowledge is stored, categorized, and searchable forever.
Available Tools
| Tool | Description |
|---|---|
nemo_save_knowledge |
Save a conversation, idea, snippet, or note |
nemo_search_knowledge |
Full-text search through your knowledge base |
nemo_get_knowledge |
Retrieve a specific entry by ID |
nemo_delete_knowledge |
Delete an entry |
nemo_list_categories |
List all categories with counts |
nemo_add_reminder |
Add a reminder with due date and priority |
nemo_list_reminders |
List pending (or all) reminders |
nemo_complete_reminder |
Mark a reminder as done |
nemo_save_bookmark |
Save a URL with tags and description |
nemo_search_bookmarks |
Search through saved bookmarks |
nemo_list_bookmarks |
List bookmarks by category |
nemo_stats |
Dashboard with counts and categories |
🏗️ Two Deployment Options
Option A: Supabase (Recommended)
Best if you want a managed database with a free tier, REST API for future apps (Flutter, web), and zero database maintenance.
Your Phone → Claude App → Your VPS (MCP Server) → Supabase Cloud (PostgreSQL)
Option B: Self-Hosted (Docker)
Best if you want full data ownership, everything on your VPS, no external dependencies.
Your Phone → Claude App → Your VPS (Docker: MCP Server + PostgreSQL)
🚀 Quick Start — Option A: Supabase
1. Create a Supabase Project
- Go to supabase.com and create a free project
- Go to SQL Editor and run the contents of
db/schema.sql - Go to Settings → API and copy your:
- Project URL (e.g.,
https://abc123.supabase.co) - Service Role Key (⚠️ keep this secret!)
- Project URL (e.g.,
2. Deploy on Your VPS
# Clone the repo
git clone https://github.com/gabriellangon/nemo-mcp.git
cd nemo-mcp
# Install dependencies
npm install
# Build
npm run build
# Configure
cp .env.example .env
# Edit .env with your Supabase credentials:
# STORAGE_TYPE=supabase
# SUPABASE_URL=https://your-project.supabase.co
# SUPABASE_SERVICE_KEY=your-service-role-key
# Test it
node dist/index.js
# Should see: 🧠 Nemo MCP server running on http://0.0.0.0:3100/mcp
3. Keep It Running with pm2
npm install -g pm2
pm2 start dist/index.js --name nemo-mcp
pm2 save
pm2 startup # Auto-start on reboot
4. Set Up HTTPS with Nginx
# Install certbot if not already done
sudo apt install certbot python3-certbot-nginx
# Get SSL certificate
sudo certbot --nginx -d nemo.yourdomain.com
# Copy nginx config
sudo cp nginx.conf /etc/nginx/sites-available/nemo-mcp
# Edit the domain name in the file
sudo ln -s /etc/nginx/sites-available/nemo-mcp /etc/nginx/sites-enabled/
sudo nginx -t && sudo systemctl reload nginx
🚀 Quick Start — Option B: Self-Hosted (Docker)
# Clone the repo
git clone https://github.com/gabriellangon/nemo-mcp.git
cd nemo-mcp
# Configure
cp .env.example .env
# Edit .env:
# STORAGE_TYPE=postgres
# DB_PASSWORD=your-secure-password
# Launch everything
docker compose up -d
# Check logs
docker compose logs -f mcp
Then set up nginx + HTTPS the same way as Option A.
🔌 Connect to Claude
Claude.ai (Web & Mobile)
- Go to Settings → Integrations → MCP
- Add a new integration:
- URL:
https://nemo.yourdomain.com/mcp
- URL:
- Start chatting and say things like:
- "Save this conversation in my brain under the category 'devops'"
- "What did I save about Docker?"
- "Add a reminder for next Friday to review the PR"
- "Bookmark this link: https://..."
Claude Desktop (Local)
Add to your claude_desktop_config.json:
{
"mcpServers": {
"nemo": {
"command": "node",
"args": ["/path/to/nemo-mcp/dist/index.js"],
"env": {
"TRANSPORT": "stdio",
"STORAGE_TYPE": "supabase",
"SUPABASE_URL": "https://your-project.supabase.co",
"SUPABASE_SERVICE_KEY": "your-service-role-key"
}
}
}
}
🔗 Webhooks — Connect to Make, Zapier, n8n
Nemo can send events to external services whenever something happens. This lets you build automations like:
- Save to Notion — Every new knowledge entry creates a page in your Notion database
- Google Calendar — New reminders automatically create calendar events
- Slack notifications — Get pinged when you save something important
- Google Sheets — Log all bookmarks to a spreadsheet
- Email digest — Trigger a daily summary via n8n
Quick Setup with Make
- In Make, create a new scenario with a Webhook trigger
- Copy the webhook URL (e.g.
https://hook.eu2.make.com/abc123...) - Add it to your
.env:WEBHOOK_URL=https://hook.eu2.make.com/abc123... - Restart the MCP server
- Now every time you save something via Claude, Make receives the event
Webhook Events
| Event | Triggered when | Data included |
|---|---|---|
knowledge.saved |
New knowledge entry saved | id, title, category, tags, content_preview, source |
knowledge.deleted |
Knowledge entry deleted | id |
reminder.created |
New reminder added | id, title, due_date, priority, description |
reminder.completed |
Reminder marked as done | id |
bookmark.saved |
New bookmark saved | id, url, title, tags, category, description |
Payload Format
Every webhook receives a JSON POST with this structure:
{
"event": "knowledge.saved",
"timestamp": "2025-03-15T14:30:00.000Z",
"data": {
"id": "uuid-here",
"title": "Docker multi-stage builds",
"category": "devops",
"tags": ["docker", "ci-cd"],
"content_preview": "First 300 characters...",
"source": "claude-chat"
}
}
Multiple Webhooks
You can send different events to different services:
WEBHOOKS_JSON='[
{
"url": "https://hook.eu2.make.com/xxx",
"events": ["knowledge.saved", "bookmark.saved"],
"name": "Make - Notion sync"
},
{
"url": "https://hooks.zapier.com/yyy",
"events": ["reminder.created"],
"name": "Zapier - Google Calendar"
},
{
"url": "https://n8n.yourdomain.com/webhook/zzz",
"events": ["*"],
"name": "n8n - Log everything"
}
]'
Security
Add a shared secret for HMAC signature verification:
WEBHOOK_SECRET=your-secret-here
Each request includes an X-Nemo-Signature header with a sha256= HMAC of the payload body. Verify this in your automation to ensure the request comes from your Nemo server.
📱 Future Ideas
- Flutter app to browse your knowledge base (connects directly to Supabase)
- Vector search with pgvector for semantic "find things similar to..."
- Image storage via Supabase Storage
- Export to Markdown/Obsidian
- Daily digest of reminders via email
🛠️ Development
# Install dependencies
npm install
# Development mode (auto-reload)
npm run dev
# Build for production
npm run build
# Run tests (coming soon)
npm test
📄 License
MIT — Use it, fork it, make it yours.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.