# Raindrop MCP Server Setup Guide
This guide explains how to connect to your deployed Raindrop MCP server at `https://web-production-46ceb.up.railway.app`.
## Quick Start
Your server is live and ready to use! Here are the connection details:
- **Server URL**: `https://web-production-46ceb.up.railway.app`
- **MCP Endpoint**: `https://web-production-46ceb.up.railway.app/mcp`
- **Health Check**: `https://web-production-46ceb.up.railway.app/health`
## Prerequisites
- A Raindrop.io account with an API access token
- The `RAINDROP_ACCESS_TOKEN` must be set in Railway (already configured)
## Connection Methods
### Method 1: Claude Desktop (Recommended)
1. **Locate your Claude Desktop config file:**
- **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
- **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`
2. **Edit the config file** and add:
```json
{
"mcpServers": {
"raindrop": {
"url": "https://web-production-46ceb.up.railway.app/mcp",
"transport": "http"
}
}
}
```
3. **Restart Claude Desktop** to load the new configuration.
4. **Verify connection**: Ask Claude to list available tools or resources from Raindrop.
### Method 2: Cursor IDE
1. **Open Cursor Settings** (Cmd/Ctrl + ,)
2. **Navigate to MCP Settings** or edit your MCP config file:
- Location: `~/.cursor/mcp.json` or in Cursor settings
3. **Add configuration**:
```json
{
"mcpServers": {
"raindrop": {
"url": "https://web-production-46ceb.up.railway.app/mcp",
"transport": "http"
}
}
}
```
4. **Restart Cursor** to apply changes.
### Method 3: MCP Inspector (Testing & Debugging)
Use the MCP Inspector to test and explore the server interactively:
```bash
npx @modelcontextprotocol/inspector https://web-production-46ceb.up.railway.app/mcp
```
This will:
- Open a web interface at `http://localhost:5173`
- Allow you to test all tools and resources
- Show real-time protocol messages
- Help debug any connection issues
### Method 4: Direct HTTP API
For programmatic access, you can make direct HTTP requests:
#### Initialize Session
```bash
curl -X POST https://web-production-46ceb.up.railway.app/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "initialize",
"params": {
"protocolVersion": "2024-11-05",
"capabilities": {},
"clientInfo": {
"name": "my-client",
"version": "1.0.0"
}
}
}'
```
#### List Available Tools
```bash
curl -X POST https://web-production-46ceb.up.railway.app/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "MCP-Session-Id: YOUR_SESSION_ID" \
-d '{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/list",
"params": {}
}'
```
**Note**: The server uses Server-Sent Events (SSE) for responses, so responses come as `event: message` followed by `data: {...}`.
## Available Tools
Once connected, you'll have access to these tools:
### Collections
- `collection_list` - List all collections
- `collection_manage` - Create, update, or delete collections
### Bookmarks
- `bookmark_search` - Search bookmarks with filters
- `bookmark_manage` - Create, update, or delete bookmarks
- `getRaindrop` - Fetch single bookmark by ID (legacy)
- `listRaindrops` - List bookmarks for collection (legacy)
- `bulk_edit_raindrops` - Bulk update bookmarks
### Tags
- `tag_manage` - Rename, merge, or delete tags
### Highlights
- `highlight_manage` - Create, update, or delete highlights
### Diagnostics
- `diagnostics` - Server diagnostic information
## Available Resources
Access these resources via MCP resource URIs:
### Static Resources
- `mcp://user/profile` - User account information
- `diagnostics://server` - Server diagnostics and environment info
### Dynamic Resources
- `mcp://collection/{id}` - Access any Raindrop collection by ID
- `mcp://raindrop/{id}` - Access any Raindrop bookmark by ID
## Testing Your Connection
### 1. Health Check
Visit in your browser:
```
https://web-production-46ceb.up.railway.app/health
```
Should return server status and active sessions.
### 2. Documentation
Visit:
```
https://web-production-46ceb.up.railway.app/
```
Shows server information and available endpoints.
### 3. MCP Inspector Test
Run:
```bash
npx @modelcontextprotocol/inspector https://web-production-46ceb.up.railway.app/mcp
```
Then in the web UI:
- Click "Initialize" to establish connection
- Browse available tools
- Test tool calls
- View resources
## Troubleshooting
### Connection Issues
1. **Verify server is running:**
```bash
curl https://web-production-46ceb.up.railway.app/health
```
2. **Check Railway logs:**
- Go to Railway dashboard
- Click on your service
- View "Deployments" → "View Logs"
3. **Verify environment variable:**
- In Railway: Settings → Variables
- Ensure `RAINDROP_ACCESS_TOKEN` is set
### Authentication Errors
If you see authentication errors:
- Verify `RAINDROP_ACCESS_TOKEN` is valid
- Check token hasn't expired (Raindrop tokens don't expire, but can be regenerated)
- Ensure token has proper permissions (read/write)
### MCP Client Not Connecting
1. **Check URL format:**
- Must use `https://` (not `http://`)
- Must include `/mcp` endpoint
- No trailing slash
2. **Verify transport type:**
- Use `"transport": "http"` for HTTP-based clients
- Some clients may use `"transport": "sse"` for Server-Sent Events
3. **Check client logs:**
- Claude Desktop: Check console output
- Cursor: Check developer console or logs
## Example Usage
### In Claude Desktop
Once connected, you can ask Claude:
- "List my Raindrop collections"
- "Search for bookmarks tagged with 'work'"
- "Create a new bookmark for https://example.com"
- "Show me highlights from my reading collection"
- "What are my most recent bookmarks?"
### In Cursor
The MCP server will be available to Cursor's AI assistant, allowing it to:
- Access your bookmarks
- Search and filter bookmarks
- Manage collections
- Work with highlights
## Advanced Configuration
### Custom Domain (Optional)
If you want a custom domain:
1. In Railway: Settings → Domains
2. Add your custom domain
3. Update MCP client configs with new URL
### Environment Variables
The server uses these environment variables (configured in Railway):
- `RAINDROP_ACCESS_TOKEN` (required) - Your Raindrop.io API token
- `PORT` (auto-set by Railway) - Server port
- `LOG_LEVEL` (optional) - Logging level (default: `info`)
- `NODE_ENV` (optional) - Environment mode (default: `production`)
## Security Notes
- The server uses your `RAINDROP_ACCESS_TOKEN` to access Raindrop.io
- All requests go through your account
- The server is publicly accessible - anyone with the URL can connect
- Consider adding authentication if you want to restrict access
- Railway provides HTTPS automatically
## Support
- **Server Status**: Check `/health` endpoint
- **Documentation**: See `README.md` for full feature list
- **Issues**: Open an issue on GitHub: https://github.com/codester1000/raindrop-mcp
## Next Steps
1. ✅ Server is deployed and running
2. ✅ Configure your MCP client (Claude Desktop, Cursor, etc.)
3. ✅ Test connection with MCP Inspector
4. ✅ Start using Raindrop tools in your AI assistant!
Enjoy your Raindrop MCP server! 🎉
