About Checklist MCP Server
Checklist MCP Server helps AI agents manage long-running work with hierarchical checklists and task state across sessions. Namespaces isolate data for separate environments, users, or projects.
Key features
- 01Create and manage nested checklists
- 02Maintain task state by session
- 03Mark tasks as completed
- 04Retrieve the full task list
- 05Isolate data with namespaces
Use cases
- 01Break long-running work into nested tasks
- 02Track an agent’s progress across sessions
- 03Separate checklists for users or projects
MCP support
The MCP server exposes checklist tools over HTTP/SSE. Clients can initialize sessions, update tasks, mark them done, and retrieve checklists.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEChecklist MCP Server
An MCP (Model Context Protocol) server for managing hierarchical checklists and tracking work context, designed to help AI agents manage long-running tasks and maintain state across sessions.
Features
- Hierarchical Task Management: Create and manage nested checklists.
- Multi-tenancy Support: Built-in support for multiple isolated environments using namespaces.
- HTTP/SSE Transport: Uses the standard MCP HTTP transport with Server-Sent Events (SSE).
Tools
| Tool Name | Description |
|---|---|
init_tasks | Initialize a new checklist session. Requires the sessionId to NOT exist. If it exists, provides steering advice with a recommended available sessionId. |
update_tasks | Create new tasks or overwrite existing tasks at a specific hierarchical path for an existing session. |
mark_task_as_done | Mark specific tasks as completed in an existing session. |
get_all_tasks | Retrieve the complete hierarchical list of tasks for an existing session. |
Session Lifecycle & Steering Guidance
- Session Initialization: Call
init_tasksfirst with a uniquesessionIdto create a checklist session. - Session Isolation: Calling
update_tasks,mark_task_as_done, orget_all_taskswith a non-existentsessionIdwill return non-error Steering guidance asking you to initialize the session first. - Conflict Prevention: Calling
init_taskswith an already existingsessionIdwill return non-error Steering guidance with a suggested alternativesessionId(e.g.my-session_1).
Quick Start
Running Locally (Node.js)
You can run the server directly using npx or npm.
Using npx:
npx checklist-mcp-server
Using npm:
git clone https://github.com/chene/checklist-mcp-server.git
cd checklist-mcp-server
npm install
npm run build
npm run start:http
The server will start on port 8585 by default.
Running with Docker
docker run -d -p 8585:8585 --name checklist-mcp-server checklist-mcp-server:latest
For more detailed Docker instructions, see DOCKER.md.
Configuration
MCP Client Configuration
Configure your MCP client (e.g., Claude Desktop, custom agent) to connect to the HTTP endpoint.
{
"mcpServers": {
"checklist": {
"transport": "http",
"url": "http://localhost:8585/mcp"
}
}
}
Multi-tenancy (Namespaces)
This server supports multi-tenancy via the namespace query parameter. This allows multiple agents or users to store distinct data within the same server instance.
To use a specific namespace, append ?namespace=<your_namespace> to the server URL.
Example Client Config for a specific namespace:
{
"mcpServers": {
"project-alpha": {
"transport": "http",
"url": "http://localhost:8585/mcp?namespace=project-alpha"
},
"project-beta": {
"transport": "http",
"url": "http://localhost:8585/mcp?namespace=project-beta"
}
}
}
If no namespace is provided, the server defaults to the default namespace.
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.