EARLY

Provides API-backed time tracking, activity management, analytics, and API access for EARLY integration.

typescript

0

GitHub Stars

typescript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "sakebomb-early_mcp": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/path/to/early_mcp",
        "python",
        "early.py"
      ],
      "env": {
        "EARLY_API_KEY": "YOUR_API_KEY",
        "EARLY_API_SECRET": "YOUR_API_SECRET"
      }
    }
  }
}

You can run an EARLY MCP Server that integrates with the EARLY Timeular v4 API to manage activities, track time, and generate analytics. This server exposes a set of tools you can invoke from your AI assistant to interact with EARLY data, making it easier to automate timekeeping and reporting.

How to use

You will connect an MCP client (such as an AI assistant) to the EARLY MCP Server and use its tools to manage activities, start and stop timers, query time entries, and generate analytics. You can start an activity, track time against it, create manual time entries, and pull summaries, efficiency reports, or billing information. Use the API explorer capabilities to make raw authenticated requests if you need to reach any EARLY API endpoint directly.

How to install

Prerequisites you must have before starting:

Python 3.12+
uv

You also need EARLY developer API credentials (an API key and secret) to access the EARLY API.

Follow these steps to install and prepare the MCP server:

git clone git@github.com:sakebomb/early_mcp.git
cd early_mcp
uv sync

Additional setup and configuration

Create environment variables to securely store your EARLY API credentials. Put these in a file named ".env" in the project directory.

EARLY_API_KEY=your_api_key
EARLY_API_SECRET=your_api_secret

Running the MCP server with Claude integration

Use the Claude Code MCP settings to point to the EARLY MCP Server. The server runs locally through the UV runtime and executes the server script that implements the tools.

uv run --directory /path/to/early_mcp python early.py

Available tools

list_activities

List all activities with IDs, names, and colors.

create_activity

Create a new activity.

update_activity

Update an existing activity's name or color.

delete_activity

Archive or delete an activity.

current_tracking

Get the currently running timer.

start_tracking

Start tracking time for an activity.

stop_tracking

Stop the current timer.

list_time_entries

Query time entries within a date range.

create_time_entry

Create a manual time entry.

delete_time_entry

Delete a time entry.

time_summary

Aggregate hours by activity for a period.

efficiency_report

Analyze time distribution and detect untracked gaps.

billing_report

Calculate billable hours and cost by activity.

explore_api

Make raw authenticated requests to any EARLY API endpoint.