Getting Started
This guide will walk you through creating your account and deploying your first MCP server on ToolBoost.
Prerequisites
Before you begin, make sure you have:
- A valid email address
- An AI client that supports MCP (Claude Desktop, Cursor, Windsurf, etc.)
Step 1: Create Your Account
- Visit toolboost.dev/sign-up
- Enter your email address and create a password
- Click "Create Account"
- Verify your email address (check your inbox)
Your account comes with:
- Free tier: 10 deployed MCPs
- 10,000 API requests per month
- Access to the entire MCP catalog
Step 2: Create a Project
After signing in, you'll be taken to your dashboard.
- Click "Create Project" or navigate to Dashboard → Projects
- Enter a project name (e.g., "My First Project")
- Optionally add a description
- Click "Create"
What is a Project?
A project is a container for your deployed MCPs. Each project has:
- Unique API Key: Used to authenticate requests
- Deployed MCPs: The MCP servers you've added
- Usage Analytics: Track requests and performance
- Team Access: Invite team members (Pro plan)
Step 3: Save Your API Key
After creating your project, you'll see your API key. Copy and save it securely - you won't be able to see the full key again.
Your API Key: tb_1234567890abcdef...
Keep your API key secret! Don't commit it to version control or share it publicly.
You can always regenerate your API key from the project settings if needed.
Step 4: Browse the MCP Catalog
Now let's find an MCP to deploy:
- Click "Catalog" in the navigation menu
- Browse through the available MCPs or use the search bar
- You can filter by category or sort by popularity
Recommended First MCPs
For beginners, we recommend starting with these popular MCPs:
- Filesystem MCP - Simple file operations, great for learning
- Memory MCP - Persistent storage for AI conversations
- Time MCP - Get current date/time information
Step 5: Deploy Your First MCP
Let's deploy the Filesystem MCP as an example:
-
Go to the Catalog
-
Search for "filesystem" or find it in the "Development" category
-
Click on the Filesystem MCP card
-
On the MCP detail page, you'll see:
- Overview and description
- Available tools
- Environment variables (if any)
- Configuration examples
-
Select your project from the dropdown in the sidebar
-
Click "Deploy Server"
Configure Environment Variables
Some MCPs require environment variables (like API keys or database URLs). If required:
- A dialog will appear showing the required variables
- Fill in the values (marked with * are required)
- Click "Attach & Save Config"
For the Filesystem MCP, you might need to configure:
ALLOWED_DIRECTORIES: Comma-separated list of directories the MCP can access
You can always update environment variables later from your project dashboard.
Step 6: Get Your Connection URL
After deployment, you'll see a success message. Now let's get your connection URL:
- Scroll to the "Use" section in the sidebar
- You'll see your Connection URL:
https://toolboost.dev/server/modelcontextprotocol/servers-filesystem/mcp?api_key=YOUR_API_KEY
- Below that, you'll find configuration examples for different clients
Step 7: Connect to Your AI Client
Choose your AI client and follow the configuration guide:
Example: Claude Desktop
For Claude Desktop, copy this configuration:
{
"mcpServers": {
"filesystem": {
"serverUrl": "https://toolboost.dev/server/modelcontextprotocol/servers-filesystem/mcp?api_key=YOUR_API_KEY"
}
}
}
Add it to your claude_desktop_config.json file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%/Claude/claude_desktop_config.json
Then restart Claude Desktop.
Step 8: Test Your MCP
Let's verify everything is working:
Using the Toolspector (Built-in Testing)
- Go back to the MCP detail page
- Click the "Toolspector" tab
- Select a tool from the list
- Fill in any required parameters
- Click "Execute" to test the tool
Using Your AI Client
Open your AI client (e.g., Claude Desktop) and try a command that uses your MCP:
For Filesystem MCP:
Can you list the files in my documents folder?
For Memory MCP:
Remember that my favorite color is blue.
If the MCP is working, your AI client will use the MCP's tools to fulfill the request!
Next Steps
Congratulations! You've successfully:
- Created your ToolBoost account
- Created a project
- Deployed your first MCP
- Connected it to your AI client
What to do next:
- Browse More MCPs - Discover more MCP servers
- Manage Environment Variables - Learn to configure MCPs
- Monitor Usage - Track your API usage
- Deploy Multiple MCPs - Add more capabilities
Troubleshooting
MCP Not Showing Up in Client
- Make sure you restarted your AI client after updating the config
- Verify your API key is correct in the connection URL
- Check that the configuration file path is correct
Environment Variable Errors
- Review required variables in the MCP detail page
- Update variables from Project Dashboard → MCPs → Configure
- Some variables may need specific formats (URLs, tokens, etc.)
Connection Errors
- Verify your internet connection
- Check that your API key hasn't been regenerated
- Ensure you're on a supported plan for the number of MCPs deployed
Get Help
Need assistance?
- Email Support: contact@toolboost.dev
- Documentation: Browse other guides in this documentation