The Claude Todo MCP Server enables persistent task management for AI coding assistants in IDEs like Cursor and Windsurf with these key capabilities:
Persistent tracking across sessions: Maintains task lists when switching projects or restarting the IDE
Multi-step task management: Breaks down complex features into manageable tasks
Single-task focus: Enforces working on one task at a time to maintain productivity
Project-specific tracking: Each workspace maintains its own todo list for context-awareness
Visual progress tracking: Displays task status (pending, in-progress, completed) clearly
Simple API: Use TodoRead to fetch the current task list and TodoWrite to update/replace it
Structured task data: Each todo contains ID, content, status, priority, and optional metadata
Automatic setup: Configurable via MCP configuration with no manual installation
Defines todo schema using TypeScript typing, enabling structured task management with proper type definitions for integration
Click on "Install 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., "@Claude Todo MCP Server show me all pending tasks for this project"
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.
Add persistent task management to any AI coding assistant in Cursor, Windsurf, and other IDEs.
This MCP server replicates Claude Code's todo system, giving your AI assistant the ability to:
Track complex multi-step tasks across sessions
Break down large features into manageable pieces
Remember progress when you switch between projects
Enforce single in-progress task focus
No installation required! Just add this to your MCP configuration:
Create .cursor/mcp.json in your project root:
{
"mcpServers": {
"task-manager": {
"command": "uvx",
"args": ["claude-todo-emulator"]
}
}
}Create config.toml at your project root:
[mcp_servers.task-manager]
command = "uvx"
args = ["claude-todo-emulator"]Add the JSON configuration above to your MCP settings file.
Restart your IDE and try: "Create a todo list for adding user authentication"
The system automatically downloads the package and sets up everything on first use!
Related MCP server: Context Management System
Once configured, your AI assistant will automatically:
Create todos for complex tasks:
You: "Add dark mode support with user preferences"
AI: "I'll create a todo list for this multi-step feature:
- ✅ Add theme context and state management
- 🔄 Create dark mode CSS variables and styles
- ⏳ Build settings UI for theme switching
- ⏳ Add persistence to localStorage
- ⏳ Update existing components for theme support"Track progress visually:
You: "What's the status on the authentication system?"
AI: "Current authentication tasks:
- ✅ Set up JWT middleware
- ✅ Create login/signup endpoints
- 🔄 Building user profile management (in progress)
- ⏳ Add password reset flow
- ⏳ Implement role-based permissions"Maintain focus:
You: "Let's also add email notifications"
AI: "I'll add that to the todo list, but let me finish the current user profile task first to maintain focus."TodoRead
Parameters: None
Returns: {todos: Todo[]}
TodoWrite
Parameters: {todos: Todo[]}
Returns: {success: boolean, count: number}
Todo Schema
{
id: string;
content: string;
status: "pending" | "in_progress" | "completed";
priority: "high" | "medium" | "low";
created_at: string;
updated_at: string;
metadata?: object;
}Fork the repository
Create a feature branch
Add tests for new functionality
Submit a pull request
MIT License - see LICENSE file for details.
Read the current task list.
Returns a list of todos with their current state.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It discloses that the tool returns a list of todos with their current state, which adds some behavioral context beyond the basic purpose. However, it lacks details on permissions, rate limits, error handling, or whether the operation is safe (e.g., read-only). For a tool with no annotations, this is a significant gap in transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise with two sentences that directly state the purpose and return value. It is front-loaded with the core function and wastes no words, making it easy for an AI agent to parse quickly. Every sentence earns its place by providing essential information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's low complexity (0 parameters, simple read operation), an output schema exists (which covers return values), and no annotations, the description is reasonably complete. It specifies what the tool does and what it returns, which is sufficient for basic understanding. However, it could benefit from more behavioral details given the lack of annotations, but the output schema mitigates this gap.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 0 parameters with 100% coverage, so no parameter documentation is needed. The description doesn't add parameter details, which is appropriate. Baseline is 4 for zero parameters, as it avoids redundancy and focuses on the tool's purpose.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose as 'Read the current task list' (verb+resource), which is specific and unambiguous. It distinguishes from the sibling TodoWrite by focusing on reading rather than writing. However, it doesn't explicitly mention what distinguishes it from potential other read operations beyond the sibling context.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context through the phrase 'current task list,' suggesting this tool is for retrieving existing todos. It doesn't provide explicit guidance on when to use this versus alternatives (though the sibling TodoWrite is clearly for writing), nor does it mention any prerequisites or exclusions. Usage is implied but not explicitly stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Update the entire task list (complete replacement).
Parameters: todos: List of todo items, each containing: - id: Unique identifier for the task - content: Task description - status: Current status (pending, in_progress, completed) - priority: Task priority (high, medium, low) - metadata: Optional additional data
Returns success status and count of todos written.
| Name | Required | Description | Default |
|---|---|---|---|
| todos | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It states this is a 'complete replacement' operation, which implies destructive behavior, but doesn't explicitly warn about data loss or confirm this is a mutation. It mentions return values but doesn't describe error handling, permissions, or rate limits. For a mutation tool with zero annotation coverage, this is insufficient.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is appropriately sized and well-structured with clear sections. The first sentence states the core purpose, followed by detailed parameter documentation and return information. Every sentence adds value, though the return statement could be slightly more concise.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given this is a mutation tool with no annotations but with output schema (returns success status and count), the description covers parameters well but lacks behavioral context. It doesn't explain what 'complete replacement' means operationally, potential side effects, or error conditions. The output schema reduces need for return value details, but overall completeness is moderate.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema has 0% description coverage, so the description must fully compensate. It provides comprehensive parameter semantics: explains 'todos' is a list, documents all nested fields (id, content, status, priority, metadata), specifies enum values for status and priority, and clarifies metadata is optional. This adds substantial meaning beyond the bare schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Update the entire task list (complete replacement).' It specifies the verb ('update'), resource ('task list'), and scope ('entire', 'complete replacement'). However, it doesn't explicitly differentiate from the sibling TodoRead tool, which prevents a score of 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus alternatives. It doesn't mention the sibling TodoRead tool, prerequisites, or scenarios where this complete replacement approach is appropriate versus incremental updates. The agent receives no usage context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v1.0.0TodoWrite1 field changedInput schema / properties / todos / titleRemoved value: -"Todos"TodoReadTodoWriteThe two tools have clearly distinct purposes: TodoRead is for retrieving the task list, while TodoWrite is for replacing it entirely. There is no overlap or ambiguity between these operations.
Both tools follow a consistent TodoVerb naming pattern (TodoRead and TodoWrite), using the same prefix and clear action verbs. This makes their functions immediately understandable and predictable.
With only two tools, the server is severely limited for a todo management domain. It lacks essential operations like creating, updating, or deleting individual tasks, making it impractical for typical todo workflows.
The tool surface is severely incomplete. It only supports reading and full replacement of the task list, missing critical CRUD operations such as add_todo, update_todo, delete_todo, or mark_complete. This will cause frequent agent failures in managing tasks.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Persistent work tracking for AI agents: tasks, status and history that follow you across machines
Adaptive plan/build/review cycles for AI coding assistants, persisted across sessions.
AI-native task management: list, create, update and archive tasks with rich context for AI agents
Gives your AI assistant persistent memory and intelligence about your work patterns.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/joehaddad2000/claude-todo-emulator'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
