todo.vu support

Help pages, tutorials and tips about how to get the most out of todo.vu

Using the todo.vu MCP server to connect your AI Agent to todo.vu

The todo.vu MCP Server enables Model Context Protocol (MCP) clients to interact directly with the todo.vu task management and time-tracking platform. It bridges the gap between AI agents and your productivity data, allowing for automated task management, time logging, and project oversight.

The server is fully hosted and requires no additional software to be installed locally. It supports the http-stream protocol and OAuth authentication.

Use Cases

• Automated Task Management: Create, update, and organize tasks based on conversational context.
• Time Tracking: Start/stop timers and log time entries directly from your chat client.
• Reporting: Fetch task statistics and project statuses to generate summaries or insights.
• Project Coordination: List clients, projects, and users to stay aligned with your team.

Tool Reference

The server currently exposes 17 tools for managing your workspace (more will be added over time):

Tasks

• list_tasks: List and filter tasks from the workspace dashboard. Supports complex filtering by client, project, user, and status.
• create_task: Create a new active task in a specific workspace.
• change_tasks: Update one or more tasks (status, name, assignment, etc.).
• repeat_task: Set or update recurrence rules (daily, weekly, etc.) for a task.
• task_stats: Get aggregated task counts and time/money statistics.
• task_add_comment: Add a rich-text or HTML comment to a task.
• list_comments_attachments_time_entries: Retrieve the full timeline of a specific task.

Time & Timers

• add_time_entry: Manually log a duration of work against a task.
• list_time_entries: Fetch and filter time logs for reporting.
• timer: Start, stop, or pause the active work timer.

Workspace & Structure

• list_workspaces: List all todo.vu workspaces your account has access to.
• list_clients: List and filter client records.
• create_client: Add a new client to the workspace.
• list_projects: List and filter project records.
• create_project: Create a new project, optionally linked to a client.
• list_labels: List available task labels/tags.
• list_users: List team members and contacts in the workspace.

Authentication & Scoping

The todo.vu MCP server supports fully automated OAuth2 authentication. If your client can support this, then you should be automatically directed to the todo.vu Authorisation page. At this point, you must be logged in to todo.vu and you just authorise access to the Full Scope.

At the moment, we only support desktop clients. More specifically, the redirect_url must be a “localhost” URL. This is the case for most desktop clients but will not work for a server-to-server connection.

Please be very careful when authorising access to your agent. While we have limited tools that are destructive, it is still possible that a rogue agent could cause significant damage to your workspace data.

Connection Details

Server URL

The primary endpoint for the MCP server is:
https://mcp.todo.vu/mcp

Authentication (OAuth 2.0)

The server uses OAuth 2.0 with PKCE. For automated clients to remain stable (avoiding frequent re-authentication),
use the following Fixed Client Identity:

• Client ID: 2e659cf8-03e6-4c14-8aa2-343a0b8d627d
• Redirect URI: http://localhost:8080/callback

Note: The port and uri may be changed to suite your client but the host must remain ‘localhost’

Configuration Samples

1. Claude Desktop

Add the following to your claude_desktop_config.json:

{ "mcpServers": { "todovu": { "url": "https://mcp.todo.vu/mcp" } } }

Note: Claude Desktop handles the OAuth flow automatically via the browser on the first run.

2. ChatGPT (via Custom Actions)

If connecting via a Custom GPT Action:

• Import from URL: https://mcp.todo.vu/mcp
• Authentication: OAuth 2.0
• Client ID: 2e659cf8-03e6-4c14-8aa2-343a0b8d627d
• Authorization URL: https://todo.vu/oauth/authorize
• Token URL: https://todo.vu/oauth/token
• Scope: full

3. Cursor / VS Code (via mcporter)

For Cursor, it is recommended to use mcporter as a bridge to ensure stable token management.

1. Run mcporter config add todovu --url https://mcp.todo.vu/mcp
2. Add the following to Cursor’s MCP settings:
   • Type: command
   • Command: mcporter call todovu

4. OpenClaw

OpenClaw uses mcporter internally for HTTP-based MCP servers. Configure it in ./config/mcporter.json:

{ "mcpServers": { "todovu": { "baseUrl": "https://mcp.todo.vu/mcp", "oauthClientId": "2e659cf8-03e6-4c14-8aa2-343a0b8d627d", "oauthRedirectUrl": "http://localhost:8080/callback" } } }

5. GitHub Copilot Extensions

If you are developing a Copilot Extension, you can point your manifest’s tool definitions to the baseUrl and use the Client ID provided above for the OAuth flow.

Share This

• Facebook
• Twitter
• LinkedIn