Temps MCP Server

The Temps MCP server implements the Model Context Protocol, letting AI assistants interact with your Temps platform directly — deploying projects, managing environments, and integrating analytics without leaving your AI coding environment.

What It Does

The MCP server exposes Temps capabilities as tools that your AI agent can call. When connected, you can say things like:

• "Deploy my current branch to production"
• "Add Temps analytics to this Next.js app"
• "Show me the logs for my API project"
• "Roll back the last deployment"

The agent uses the MCP tools to execute these operations directly against your Temps instance.

Installation

Claude Desktop

Add the Temps MCP server to your claude_desktop_config.json:

{
  "mcpServers": {
    "temps": {
      "command": "npx",
      "args": ["@temps-sdk/mcp"],
      "env": {
        "TEMPS_API_URL": "https://your-temps-instance.com",
        "TEMPS_API_TOKEN": "your-api-token"
      }
    }
  }
}

Config file location:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Claude Code

npx skills add gotempsh/temps --skill temps-mcp-setup

This installs the temps-mcp-setup skill, which guides your AI agent through the full MCP configuration interactively.

Cursor / Windsurf / Other MCP Clients

{
  "mcp": {
    "servers": {
      "temps": {
        "command": "npx",
        "args": ["@temps-sdk/mcp"],
        "env": {
          "TEMPS_API_URL": "https://your-temps-instance.com",
          "TEMPS_API_TOKEN": "your-api-token"
        }
      }
    }
  }
}

Getting Your API Token

1. Open your Temps dashboard
2. Go to Settings → API Tokens
3. Click New Token and give it a name
4. Copy the token — it won't be shown again

Or via CLI:

bunx @temps-sdk/cli tokens create --name "mcp-local"

Managing MCP Servers via CLI

The mcp-servers command manages MCP server registrations on your Temps instance:

# List registered MCP servers
bunx @temps-sdk/cli mcp-servers list

# Add an MCP server (--config accepts the MCP server config as JSON)
bunx @temps-sdk/cli mcp-servers add --name my-server --config '{"url":"https://my-mcp.example.com"}'

# Remove an MCP server (pass the slug/name as a positional argument)
bunx @temps-sdk/cli mcp-servers delete my-server --yes

Available Prompts

The Temps MCP server exposes the following prompts your AI agent can invoke:

add_react_analytics

Comprehensive guide to adding Temps analytics to a React application.

Arguments:

Covers: installation, provider setup, custom events, scroll tracking, session recording, pageview tracking, and user identification.

Related