> ## Documentation Index > Fetch the complete documentation index at: https://docs.taketime.space/llms.txt > Use this file to discover all available pages before exploring further. # Take Time MCP Examples — Natural Language Prompts > Practical natural-language prompt examples for managing your Take Time schedule with Claude, Cursor, or any MCP-compatible AI assistant. Once the Take Time MCP Server is installed, your AI assistant can help you manage your entire study schedule through conversation. Here are practical examples of what you can ask Claude or Cursor to do — and what happens behind the scenes when you do. *** ## Reviewing Your Day These prompts help you get a quick picture of what's on your plate and where you stand. ### "What study blocks do I have today?" **What the AI does:** Calls `list_blocks` with `date` set to today's date. Returns a formatted list of all blocks for the day, including subject names, times, topics, and whether each one has been completed. *** ### "Which blocks haven't I completed yet?" **What the AI does:** Calls `list_blocks` with `date` set to today and `done: false`. Surfaces only incomplete sessions so you can focus on what's still left to do. *** ### "How many hours am I scheduled to study this week?" **What the AI does:** Calls `get_schedule` with `week_offset: 0`, then totals the duration of all blocks across the week's days. Returns the sum in hours and minutes, broken down by day if you ask. *** ## Creating and Editing Blocks These prompts let you build and adjust your schedule without touching the app. ### "Schedule a JavaScript block tomorrow from 9am to 10am" **What the AI does:** Calls `list_subjects` to find the subject ID that matches "JavaScript", then calls `create_block` with the appropriate `subject_id`, tomorrow's date, `start: "09:00"`, and `end: "10:00"`. Confirms the created block and its ID. *** ### "Mark my morning block as done" **What the AI does:** Calls `list_blocks` with today's date to identify your earliest block, then calls `update_block` with that block's `id` and `done: true`. Confirms the update and shows your revised completion count for the day. *** ### "Move my Python session to 3pm" **What the AI does:** Identifies your Python block for today using `list_blocks` filtered by subject, then calls `update_block` to update `start` and `end` to `"15:00"` and the appropriate end time. The duration stays the same unless you specify a new one. *** ## Checking Progress Use these prompts to track your momentum and hold yourself accountable. ### "What's my completion rate this week?" **What the AI does:** Calls `get_stats` with `period: "week"` and returns your `completion_rate` as a percentage, along with the raw counts (`completed_blocks` out of `total_blocks`). *** ### "How long is my current study streak?" **What the AI does:** Calls `get_stats` with `period: "all"` and surfaces the `streak_days` value. If your streak has reset recently, the AI can also pull last week's stats to help you understand what broke it. *** ### "Show me my stats for the past month" **What the AI does:** Calls `get_stats` with `period: "month"` and presents a summary covering your `total_minutes`, `completion_rate`, `streak_days`, and a description of your heatmap activity — highlighting your most and least productive days. *** ## Schedule Planning These prompts use the AI's reasoning to build and improve your weekly plan. ### "Suggest a better schedule for me — I want to study more math" **What the AI does:** Calls `suggest_schedule` with `goal: "study more math"`. Returns a set of recommendations explaining which days and times are best suited for math based on your past completion patterns. You review the suggestions and confirm which ones to apply. *** ### "Add a React activity to my schedule on Monday, Wednesday, and Friday from 2–3pm" **What the AI does:** Calls `list_subjects` to locate (or confirm) a React subject, then calls `create_block` three times — once per day — with `start: "14:00"` and `end: "15:00"`. Summarizes all three created blocks at once. *** You can chain multiple requests in a single message. For example: *"List my incomplete blocks for today, then mark all of them as done."* The AI will call `list_blocks` first, then call `update_block` in sequence for each result — reporting back after every step so you stay in control.