eranfinish/ezatwork-mcp

eranfinish/ezatwork-mcp

MCP server for EZ@Work — manage clients, projects, time entries, and invoices from Claude through natural conversation. OAuth 2.1 with Google sign-in, multi-currency support.

Category
Visit Server

README

EZ@Work MCP Server

Connect EZ@Work — all-in-one business management for freelancers, lawyers, agencies, and service businesses — to Claude, Gemini, Cursor, and any MCP-compatible AI.

Tools (v1)

Tool Description Type
list_clients List your clients/customers Read
list_projects List your projects/cases/work orders Read
log_time_entry Log billable hours to a project Write
create_invoice Create a draft invoice (auto currency/language/VAT) Write

Setup

Claude Desktop / Claude.ai (OAuth — recommended)

Settings → Connectors → Add custom connector

  • URL: https://mcp.ezatwork.com/mcp
  • Authentication: OAuth (automatic)

Claude Code / Cursor / MCP Inspector (API token)

  1. Create a free EZ@Work account at https://app.ezatwork.com
  2. Generate an API token at https://app.ezatwork.com/settings/api-tokens
  3. Connect to https://mcp.ezatwork.com/mcp with header Authorization: Bearer ezw_pat_...

Local testing with MCP Inspector:

npx @modelcontextprotocol/inspector

Connect to http://localhost:8080/mcp with header Authorization: Bearer ezw_pat_...

Security

  • Your data stays in your EZ@Work account
  • OAuth mode: Google identity → EZ@Work account lookup; no credentials stored in the MCP
  • API token mode: scoped permissions — grant only what you need; mandatory expiration
  • Authorization is verified on every request

Privacy Policy

https://www.ezatwork.com/privacy

Universal Design

EZ@Work adapts to your business type: a lawyer sees "cases," a freelancer sees "projects," a service business sees "work orders." Currency, language (24 supported), and tax rules auto-detect from your profile — no configuration needed in the MCP.


Development

Run locally

python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -e .

cp .env.example .env
# Edit .env: EZ_API_BASE=http://localhost:5000 for local backend

python -m ezatwork_mcp.server
# Server starts at http://localhost:8080/mcp

Run tests

pip install pytest pytest-asyncio
pytest tests/

Docker

docker build -t ezatwork-mcp .
docker run -p 8080:8080 -e EZ_API_BASE=https://api.ezatwork.com ezatwork-mcp

Deploy to Cloud Run

gcloud run deploy ezatwork-mcp \
  --source . \
  --region europe-west1 \
  --project ezatwork-production \
  --allow-unauthenticated \
  --port 8080 \
  --update-env-vars "EZ_API_BASE=https://api.ezatwork.com,MCP_BASE_URL=https://mcp.ezatwork.com" \
  --update-secrets "GOOGLE_CLIENT_ID=ezmcp-google-client-id:latest" \
  --update-secrets "GOOGLE_CLIENT_SECRET=ezmcp-google-client-secret:latest" \
  --update-secrets "EZ_MCP_MASTER_TOKEN=ezmcp-master-token:latest"

--allow-unauthenticated is correct — auth is enforced at the MCP layer (OAuth or API token), not at the Cloud Run IAM level.

Technical notes

  • Transport: Streamable HTTP (/mcp endpoint)
  • Library: fastmcp 3.3.1
  • Auth (OAuth): GoogleProvider from fastmcp.server.auth.providers.google
  • Auth (token): get_http_headers() from fastmcp.server.dependencies
  • Dual-auth: OAuth active when GOOGLE_CLIENT_ID + GOOGLE_CLIENT_SECRET + MCP_BASE_URL are set
  • Backend: GET /api/users/me for locale; GET /api/internal/users/by-email for OAuth→user mapping
  • Time entries: POST /api/timeentry with isBillable field
  • Invoices: POST /api/invoices with computed dueDate and total per item

Built by

EZ@Work — the business OS for independents.

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured