How to use slog-app MCP ?

What Can You Do with Slog’s MCP Server?

With Slog’s MCP server, your AI assistant becomes your personal time tracking and project management assistant. You can ask it to:

Track Time – Start and stop time entries for specific projects and clients
View Time Entries – Check your logged hours, today’s work, or time for specific projects
Manage Projects – Create new projects, view existing ones, and get project details
Handle Clients – View and manage client information
Team Management – View team members, working hours, and schedules

Using Slog MCP with Cursor

Cursor is a powerful AI-powered code editor that supports MCP. Here’s how to set it up:

Step 1: Get Your API Credentials

First, you’ll need your Slog API token. Get it from your account settings.

Step 2: Configure Cursor

In Cursor, you need to edit the MCP configuration file. On macOS, this is typically located at:

~/.cursor/mcp.json

Create this file if it doesn’t exist, or edit the existing one.

Step 3: Add Slog Configuration

Add the following configuration to your mcp.json file:

{
  "mcpServers": {
    "slog": {
      "url": "https://www.slog-app.com/mcp",
      "type": "streamablehttp",
      "headers": {
        "Authorization": "Bearer YOUR_SLOG_API_TOKEN"
      }
    }
  }
}

Important: Replace YOUR_SLOG_API_TOKEN with your actual API token from Step 1.

Step 4: Restart Cursor

After saving the configuration file, restart Cursor to load the new MCP server.

Configuration for Other AI Assistants

While this guide focuses on Cursor, Slog’s MCP server works with any MCP-compatible AI assistant. The configuration format is similar for most clients:

Claude Desktop – Edit claude_desktop_config.json
Continue.dev – Edit ~/.continue/config.json
Other MCP clients – Refer to your client’s documentation

Using Slog MCP with Developer Tools

Beyond AI assistants, you can also integrate Slog’s MCP server programmatically using developer tools and libraries. This is perfect for building custom integrations, automated workflows, or embedding Slog functionality into your own applications.

Ruby Integration with ruby_llm-mcp

For Ruby developers, you can use the ruby_llm-mcp gem to connect to Slog’s MCP server. This library provides full MCP protocol support and integrates seamlessly with RubyLLM.

Example Setup:

require 'ruby_llm/mcp'

# Connect to Slog's MCP server
client = RubyLLM::MCP.client(
  name: "slog",
  transport_type: :streamable,
  config: {
    url: "https://www.slog-app.com/mcp",
    headers: {
      "Authorization" => "Bearer YOUR_SLOG_API_TOKEN"
    }
  }
)

# Get available tools
tools = client.tools
puts "Available Slog tools:"
tools.each do |tool|
  puts "- #{tool.name}: #{tool.description}"
end

# Use tools with RubyLLM chat
chat = RubyLLM.chat(model: "gpt-4")
chat.with_tools(*client.tools)

# Ask questions that use Slog's MCP tools
response = chat.ask("Track time for the 'API Development' project")
puts response

The ruby_llm-mcp gem supports multiple transport types including SSE (Server-Sent Events), STDIO, and streamable HTTP. It’s perfect for building AI-powered applications that can seamlessly interact with your time tracking data.

Other Integration Options

Since Slog’s MCP server adheres to the Model Context Protocol specification, you can use any MCP-compatible library in your preferred programming language:

Other languages – Any language that can make HTTP requests can interact with the SSE endpoint

Python – Use the official MCP Python SDK or libraries like mcp-client

JavaScript/TypeScript – Use the official @modelcontextprotocol/sdk package

Documentation