get_projectsLABSCO SUMMARY
Read-only as a design decision, not a gap to work around.
There is no write path at all, which makes this a connection you can leave open without thinking about it — the worst outcome is an assistant reading a project it did not need. The action-required view is the most useful tool in the set: it answers what is waiting on you directly, rather than making you filter an assignment list.
What it is
A read-only server over the Goodday project management API. It lists projects and their members, reads tasks by project or by the person they belong to, flags the ones needing action, and looks up users.
What you get
- Projects listed, with archived and root-only filters, and one project read in full
- The users on a project
- A project's tasks, optionally including closed ones and subfolders
- The tasks assigned to a person, and separately the ones requiring their action
- One task read in full
- The organization's users, and one user read in full
- The same tools over a local HTTP server with generated API documentation, when you would rather not run it as a subprocess
Requirements
A Goodday API token from your account settings, supplied as an environment setting. Python 3.10 or newer with uv, running from a clone of the repository; the Docker image is one you build yourself from its Dockerfile. Every request is a read — the server has no path to create, update or delete anything.
Setup effort
Build from source — git clone https://github.com/cdmx-in/goodday-ro-mcp.git
USE CASES
01
Asking what is waiting on you
Read the action-required list for a person, which is a different question from what they have been assigned.
02
Getting a project's status
Read the project and its open tasks, pulling in closed ones and subfolders only when you want the full picture.
03
Finding out who is on a project
List the project's members, and resolve any of them against the organization's user list.
04
Checking someone's workload
Read what is assigned to them across projects, rather than opening each project in turn.
05
Reading the detail behind a task
Pull one task in full when the summary in a list is not enough to answer the question.
9 tools
Listed in project docs. Your client may show a different list.
get_projectget_project_usersget_project_tasksget_taskget_user_assigned_tasksget_user_action_required_tasksget_usersget_userBEFORE YOU INSTALL
What to know first
Tools you get
9 tools named in the project's documentation.
Requires
- Python 3.10 or later
- uv
- GOODDAY_API_TOKEN in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
git clone https://github.com/cdmx-in/goodday-ro-mcp.git
cd goodday-ro-mcp
uv syncMCP Clients
Claude Desktop
Runs the goodday-ro-mcp script from your clone (uv sync installed it into the project environment). <YOUR_PATH_TO_GOODDAY_RO_MCP> is the folder you cloned into and put your Goodday API token in GOODDAY_API_TOKEN.
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"goodday-ro": {
"command": "uv",
"args": [
"--directory",
"<YOUR_PATH_TO_GOODDAY_RO_MCP>",
"run",
"goodday-ro-mcp"
],
"env": {
"GOODDAY_API_TOKEN": "<YOUR_GOODDAY_API_TOKEN>"
}
}
}
}
