Setup: VS Code
Connect VS Code to ProBeya via the Claude MCP extension — configure the server in settings.json and start using tools inline.
Prerequisites
- VS Code 1.96+ installed
- The Claude extension for VS Code installed
- A ProBeya API key (generate from Settings > API Keys in the ProBeya web app)
- Node.js 18+ installed
- The ProBeya MCP server built (
apps/mcp/dist/index.js)
Step 1: Install the Claude Extension
Open VS Code and install the Claude extension from the marketplace:
- Open the Extensions panel (
Ctrl+Shift+X/Cmd+Shift+X) - Search for “Claude” by Anthropic
- Click Install
Or install from the command line:
code --install-extension anthropic.claude-code
Step 2: Configure the MCP Server
Add the ProBeya MCP server to your VS Code settings. Open your settings JSON file:
- Open the Command Palette (
Ctrl+Shift+P/Cmd+Shift+P) - Type “Preferences: Open User Settings (JSON)”
- Press Enter
Add the MCP server configuration:
{
"claude.mcpServers": {
"probeya": {
"command": "node",
"args": ["/path/to/probeya/apps/mcp/dist/index.js"],
"env": {
"PRPROBEYA_API_KEY": "probeya_sk_live_...",
"DATABASE_URL": "postgresql://user:pass@host:5432/probeya"
}
}
}
}
Replace /path/to/probeya/ with the actual absolute path to your ProBeya installation.
Replace the placeholder values for PRPROBEYA_API_KEY and DATABASE_URL with your real credentials.
Workspace-Level Configuration
For team projects, you can add the configuration to .vscode/settings.json in
your workspace root. This way, all team members sharing the repository get the
MCP server preconfigured (they still need to provide their own env variables):
{
"claude.mcpServers": {
"probeya": {
"command": "node",
"args": ["${workspaceFolder}/apps/mcp/dist/index.js"],
"env": {
"PRPROBEYA_API_KEY": "${env:PRPROBEYA_API_KEY}",
"DATABASE_URL": "${env:DATABASE_URL}"
}
}
}
}
The ${env:VARIABLE} syntax reads from the developer’s system environment variables,
keeping secrets out of version control.
Step 3: Reload VS Code
After saving the settings, reload VS Code to pick up the MCP server configuration:
- Open the Command Palette (
Ctrl+Shift+P/Cmd+Shift+P) - Type “Developer: Reload Window”
- Press Enter
Step 4: Test the Connection
Open the Claude panel in VS Code and try a simple query:
“List my ProBeya workspaces”
Claude should invoke the list_workspaces tool and display your organization’s
workspace hierarchy.
You can verify the MCP server started successfully by checking the Output panel:
- Open the Output panel (
Ctrl+Shift+U/Cmd+Shift+U) - Select “Claude” from the dropdown
- Look for the ProBeya MCP server startup messages:
[MCP] Authenticating...
[MCP] Jane Smith ([email protected]) -- org org_abc123
[MCP] Scopes: full access
[MCP] ProBeya MCP server v2.0.0 ready (stdio)
Common Workflows in VS Code
Inline Board Context
While editing code, ask Claude about your ProBeya boards:
“What items are in the Quality Improvement project?”
Claude will call list_workspaces, then list_projects, then get_board to
navigate to the right data.
Creating Items from Code Comments
When reviewing code with TODO comments:
“Create a ProBeya item for each TODO comment in this file”
Claude can use create_item to create board items directly from your code context.
KPI Monitoring During Development
Check if your changes impacted KPIs:
“Show me KPI alerts for the production board”
Claude will call get_kpi_alerts and display any red or amber threshold breaches.
Troubleshooting
Security Notes
- When using workspace-level settings (
.vscode/settings.json), always use${env:VARIABLE}references for secrets — never commit raw API keys - The MCP server runs as a local child process spawned by VS Code, communicating
over stdio — data does not traverse external networks (unless your
DATABASE_URLpoints to a remote host) - Consider using VS Code’s built-in secret storage for managing sensitive environment variables across projects
Was this page helpful?