Connecting with AI assistants (MCP)
If you use an AI assistant like Claude or ChatGPT for daily work, you can now connect it to Everhour and let it track time, log entries, and read your timesheet on your behalf. No more switching tabs to start a timer — just ask.
In this article:
How to setup
Everhour exposes an MCP (Model Context Protocol) server at https://api.everhour.com/mcp. Any AI client that supports MCP can connect to it with your Everhour API key.
You'll need:
- An active Everhour account.
- Your Everhour API key — copy it from Everhour → My Profile → API key
- One of the supported clients below.
Pick the client you already use; setup takes only a couple of minutes.
Claude Code
Claude Code is an agentic coding tool that reads your codebase, edits files, runs commands, and integrates with your development tools. Available in your terminal, IDE, desktop app, and browser. It supports remote MCP servers natively, so connecting is a single command.
If Claude Code is already installed
- Open your terminal and run the following command, with your API key:
claude mcp add --transport http everhour https://api.everhour.com/mcp --header "x-api-key: YOUR_EVERHOUR_API_KEY"
- Then in Claude Code, type
/mcpand confirm that everhour is listed and shows as connected.
If Claude Code is not installed
You can read full setup guide following Claude Code's doc. After installation, run:
claude mcp add --transport http everhour https://api.everhour.com/mcp --header "x-api-key: YOUR_EVERHOUR_API_KEY"
Then type /mcp in Claude Code to verify the connection.
Cursor
Cursor reads MCP servers from a config file:
- Open (or create) the file
~/.cursor/mcp.jsonand add themcpServersblock with the Everhour URL and your API key header. - Restart Cursor.
- Open Cursor Settings → MCP and check that
everhourshows a green status indicator.
Claude Desktop
Claude Desktop connects to MCP servers via stdio, while Everhour MCP uses HTTP. To bridge this, a small helper ( mcp-remote) is used.
Setup steps
- Open Claude Desktop → Settings → Developer → Edit Config
- Replace or merge your config with the following:
{
"mcpServers": {
"everhour": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://api.everhour.com/mcp",
"--header",
"x-api-key: YOUR_EVERHOUR_API_KEY"
]
}
}
}
Final steps
- Save the file
- Fully quit Claude Desktop (not just close the window)
- Reopen Claude Desktop
Optional
If your firewall blocks npx, install the bridge globally:
npm install -g mcp-remote
Then update the config:
{
"mcpServers": {
"everhour": {
"command": "mcp-remote",
"args": [
"https://api.everhour.com/mcp",
"--header",
"x-api-key: YOUR_EVERHOUR_API_KEY"
]
}
}
}
ChatGPT
ChatGPT connects to Everhour through its Connectors feature.
- Open ChatGPT → Settings → Connectors → Add custom connector.
- Fill in the Server URL (
https://api.everhour.com/mcp), set authentication to custom header, and enterx-api-key. - Save the connector and enable it.
Other clients
Any client that supports MCP works with the same URL and header: https://api.everhour.com/mcp + x-api-key.
Available tools
Once connected, your AI gets access to these 20 tools:
| Area | Tools |
| Timer | everhour_get_current_timer, everhour_start_timer, everhour_stop_timer |
| Time entries | everhour_log_time, everhour_update_time_entry, everhour_delete_time_entry, everhour_get_user_time, everhour_get_my_timesheet |
| Tasks | everhour_search_tasks, everhour_recent_tasks, everhour_get_task, everhour_list_tasks, everhour_resolve_platform_ids |
| Projects & clients | everhour_list_projects, everhour_get_project, everhour_list_clients, everhour_get_client |
| Team | everhour_get_me, everhour_list_team_users |
| Platforms | everhour_list_platforms |