home / mcp / todoist mcp server
Provides Todoist task management capabilities to AI assistants via an MCP server over stdio.
Configuration
View docs{
"mcpServers": {
"bkotos-todoist-mcp": {
"command": "$PATH_TO_REPO/start-mcp.sh",
"args": []
}
}
}This TypeScript-based MCP server provides Todoist task management capabilities to AI assistants, operating over stdio to enable seamless integration with Claude Desktop. It helps you manage Todoist projects and tasks through firmware-like MCP interactions, empowering your AI workflows with real-time task data and actions.
You connect to the MCP server from your MCP client when you want to perform Todoist operations. Start the local MCP server using the configured startup script, then configure your MPC client to communicate with it over stdio. Once connected, you can request lists of Todoist projects and tasks, create new tasks, update task details, and fetch project structures as part of your AI-assisted workflows. Use the same commands the client provides to invoke common Todoist actions, and rely on the server to translate those requests into Todoist API calls.
Prerequisites include a modern Node.js runtime and access to Todoist API tokens.
git clone <repository-url>
cd todoist-mcp
npm ciConfigure your environment by copying the example env file and adding your Todoist API token.
cp env.example .env
# Add your Todoist API token to .envSet up the Claude Desktop integration to discover and connect to your local MCP server. You will point Claude Desktop to the startup script that launches the MCP server.
Below is the exact MCP server configuration snippet you should place in Claude Desktop so it can launch your local server.
{
"mcpServers": {
"todoist": {
"command": "$PATH_TO_REPO/start-mcp.sh",
"args": [],
"env": {}
}
}
}Restart Claude Desktop after updating the MCP configuration. Then test the integration by asking Claude to list your Todoist projects or tasks to confirm the server responds with your Todoist data.