Skip to main content
The Model Context Protocol (MCP) is an open standard that lets AI tools connect to external services. Znift hosts an MCP server that gives your AI agent full access to your tasks.

​
Configuration

Add Znift to your AI tool’s MCP configuration. Replace your-znift-api-key with your actual API key from Profile > API Keys.
Edit your Claude Desktop config file:
  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "znift": {
      "url": "https://znift.com/mcp",
      "headers": {
        "x-api-key": "your-znift-api-key"
      }
    }
  }
}
Restart Claude Desktop after saving.

​
Available Tools

Once connected, your AI agent has access to the following tools:
ToolDescription
list_tasksList all tasks for a specific date
create_taskCreate a new task with text, date, priority, and more
get_taskGet a single task by ID
update_taskUpdate fields on an existing task
delete_taskPermanently delete an archived task
search_tasksFull-text search across all tasks
get_analyticsCompletion stats and trends for a date range
get_dashboardDaily overview with overdue, pending, completed, and streak
list_archivedList all archived tasks
archive_taskMove a task to the archive
restore_taskRestore an archived task to its original date
toggle_taskToggle a task between completed and pending
move_taskMove a task to a different date

​
Example Conversations

After setup, try these prompts with your AI agent:
  • “What’s on my plate today?”
  • “Add a high-priority task for tomorrow: finalize the proposal”
  • “Log that I merged 3 PRs today”
  • “Record that I had a 1:1 with Sarah yesterday”
  • “Mark the code review task as done”
  • “Move the team sync to next Monday”
  • “Search for anything about the API migration”
  • “How productive was I this week?”

​
Troubleshooting

Make sure you’ve restarted your AI tool after adding the MCP configuration. Check that your API key is correct and starts with znift_.
Verify your API key at Profile > API Keys. Generate a new key if needed.
The API allows 300 requests per minute per API key. If you hit the limit, wait a moment and try again.