Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-server-interview Create a new todo list called 'Work'."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Model Context Protocol server for croonchloop interview
The following steps were how the server was tested.
First a client like claude desktop should be installed. Then open a terminal in the mcp-server-interview folder and execute:
Add MCP to your project dependencies:
uv add "mcp[cli]"
Then install the mcp server in claude:
uv run mcp install src/main.py
Related MCP server: Todoist MCP Server
MCP-SERVER-INTERVIEW/
├── requirements.txt
├── .gitignore
├── config/
│ ├── conf.yml
| ├── loader.py
│ └── __init__.py
├── src/
| ├── dto/
| │ ├── lists.py
| │ └── tasks.py
| ├── service/
| | ├── __init__.py
| │ ├── todoList.py
| │ └── tasks.py
│ └── main.py
├── README.md
└── requirements.txtExamples of propts given to claude desktop to test and interact with the server. The prompts were given in spanish like the example of the challenge instructions.
Create a new todo list.
Crea una lista llamada 'Nombre de lista"
Create a task for a specific list.
Crea la tarea llamada 'tarea1' con descripcion 'desc' en la lista llamda 'Nombre de lista'
Start a task (change the task's status to 'InProgress')
Comenzar la tarea 'tarea1' de la lista 'Nombre de lista'
Complete a task (change the task's status to 'Completed')
Terminar la tarea 'tarea1' de la lista 'Nombre de lista'
Delete a task
Eliminar la tarea 'tarea1' de la lista 'Nombre de lista'
Tasks: Right now it's not avilable to change name and description for a giving task.
Test: Even though the server was manually tested, testing the service layer will ensure correctness in futures changes.
App Configuration: After having dependency-related errors, the app configuration module was left out, and the todo list service url was hardcoded in the main script. However, keeping all configuration values in one place make te deployment and usage easier.
Project structure: Structuring a project in layers helps to assign responsabilities and group similar functions and classes. For this kind of project, it's not common to find a architecture design to follow.
Complete, terminate o finalizar una tarea / task con nombre 'name' en la lista llamada 'listname'
Inputs: list (required), name (required).
Crear lista ‘Prueba’
Inputs: name (required).
Crear task o tarea ‘Prueba’ con descripcion 'desc' en la lista 'nombre de lista'
Inputs: desc (required), list (required), name (required).
Eliminar, remover o quitar una tarea / task con nombre 'name' de la lista llamada 'listname'
Inputs: list (required), name (required).
Obter, ver la lista con nombre ‘Prueba’
Inputs: name (required).
Comenzar, empezar o start una tarea / task con nombre 'name' en la lista llamada 'listname'
Inputs: list (required), name (required).
Cambiar, actualizar o update el nombre de la lista ‘OldName’ a o por 'NewName'
Inputs: newName (required), oldName (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 7 tools
All seven tools target distinct operations: list creation, list renaming, list retrieval, task creation, task starting, task completion, and task deletion. No two tools have overlapping purposes.
Tool names follow a verb_noun pattern, but there is inconsistency: 'start_a_task' and 'complete_a_task' include an extra 'a_', while others like 'create_task' and 'delete_a_task' vary in article usage. This breaks the otherwise consistent pattern.
Seven tools cover the essential operations for a list-based task manager without being overly numerous or insufficient. The count feels right for the domain.
Basic CRUD is present but incomplete: there is no delete list, no update task (e.g., description or name), and no way to list all tasks in a list. Notable gaps exist for a full task management workflow.
AI-powered to-do list: add, list, complete and organise tasks across lists from any assistant.
To-do lists by voice or chat: add tasks with due dates, check them off, see what's due.
Manage Superlist tasks and lists in plain language from any MCP-compatible AI agent.
Create, list, and complete todo items through MCP.