MCP.so
Sign In

Harvest Time Tracking MCP Server

@mayank2424

RepositoryHomepage

About Harvest Time Tracking MCP Server

MCP (Model Context Protocol) server for Harvest Time Tracking.

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "Harvest": {
      "command": "npx",
      "args": [
        "@harvest/mcp-server-typescript"
      ],
      "env": {
        "HARVEST_ACCOUNT_ID": "<Harvest Account ID>",
        "HARVEST_ACCESS_TOKEN": "<Harvest Personal Access Token>",
        "HARVEST_BASE_URL": "<Harvest Base URL>"
      }
    }
  }
}

Tools

12

Get company information

Searches for clients

Get client information

Searches for projects

Get project information

Lists all users

Get user information

List project user assignments

List project task assignments

Searches for time entries

Get time entry details

Create a new time entry

Overview

What is Harvest Time Tracking MCP Server?

A TypeScript-based MCP (Model Context Protocol) server that enables LLM clients to interact with a Harvest time tracking account. It provides tools to query and manage Harvest data such as clients, projects, users, time entries, and company information.

How to use Harvest Time Tracking MCP Server?

Install via npx -y @mayankg0324/mcp-server-harvest. Configure the environment variables HARVEST_ACCOUNT_ID, HARVEST_ACCESS_TOKEN, and HARVEST_BASE_URL in your MCP client’s config file (e.g., Claude Desktop claude-desktop-config.json or Cursor IDE mcp.json). Then use the available tools (e.g., search-clients, create-time-entry) through your LLM client.

Key features of Harvest Time Tracking MCP Server

  • Retrieve company information (name, URL, ID, currency)
  • Search and get details for clients, projects, and users
  • List project user assignments and task assignments
  • Search and fetch time entry details with filters
  • Create new time entries with project, task, user, and hours
  • Works with Claude Desktop and Cursor IDE out of the box

Use cases of Harvest Time Tracking MCP Server

  • Ask an LLM to find all active clients in your Harvest account
  • Query which users are assigned to a specific project
  • Retrieve a summary of time entries for a given date range
  • Create a new time entry for a specific project and task through natural language
  • List all project task assignments to understand workload

FAQ from Harvest Time Tracking MCP Server

What are the prerequisites for using Harvest Time Tracking MCP Server?

You need Node.js >= 18 (or Bun for development), a Harvest account with API access, a Personal Access Token from Harvest, and your Harvest Account ID.

What environment variables are required?

The server requires three environment variables: HARVEST_ACCOUNT_ID, HARVEST_ACCESS_TOKEN, and HARVEST_BASE_URL.

How do I fix spawn npx ENOENT or env: node: No such file or directory?

This error occurs when your MCP client cannot find npx or node in its PATH. Use the full path to npx (found via which npx) and include your Node bin directory in the PATH environment variable inside the config.

Which MCP clients are supported?

The README explicitly provides configuration examples for Claude Desktop and Cursor IDE. Other MCP clients may work but are not documented.

Can I run the server locally for development?

Yes. Clone the repository, run npm install and npm run build, then point your MCP client config to the local dist/index.js using the node command.

Frequently asked questions

What are the prerequisites for using Harvest Time Tracking MCP Server?

You need Node.js >= 18 (or Bun for development), a Harvest account with API access, a Personal Access Token from Harvest, and your Harvest Account ID.

What environment variables are required?

The server requires three environment variables: `HARVEST_ACCOUNT_ID`, `HARVEST_ACCESS_TOKEN`, and `HARVEST_BASE_URL`.

How do I fix `spawn npx ENOENT` or `env: node: No such file or directory`?

This error occurs when your MCP client cannot find `npx` or `node` in its PATH. Use the full path to `npx` (found via `which npx`) and include your Node bin directory in the `PATH` environment variable inside the config.

Which MCP clients are supported?

The README explicitly provides configuration examples for Claude Desktop and Cursor IDE. Other MCP clients may work but are not documented.

Can I run the server locally for development?

Yes. Clone the repository, run `npm install` and `npm run build`, then point your MCP client config to the local `dist/index.js` using the `node` command.

Comments

Basic information

Category

Productivity

License

MIT

Transports

stdio

Author

mayank2424

More Productivity MCP servers

Nevent

nevent

Nevent is a remote MCP server available at https://mcp.nevent.ai.

Etincel

AIStoryHub

Find the AI tells in your prose. Deterministic, local, runs in CI. MCP server + CLI + Action.

6

FlowyTeam OKR MCP Server

flowy-team

FlowyTeam is a native OKR MCP server. Connect Claude (Desktop, web, mobile), ChatGPT, Claude Code or n8n and your agent can read every objective and key result in your workspace, post check-ins, move key-result progress,

AdWhispr - Research, Clone, & Launch Ads from Claude/ChatGPT

AdWhispr

Remote MCP server that turns Claude, ChatGPT, or Cursor into a full ad workflow: research any brand's live Facebook and TikTok ads, clone the proven winners for your own brand, and launch real campaigns on Google, TikTok

LinkedIn Ghostwriter

ContentIn

Write LinkedIn posts in your own voice from Claude, ChatGPT, or any MCP client. ContentIn trains a personal AI on your own LinkedIn posts, then generates ideas, drafts in your voice, schedules, publishes via LinkedIn's o