Study Planner MCP

Study Planner MCP

Offline exam planning for your AI: no account, no login, no internet needed, nothing stored. Free forever, open source.

Give it your exam dates and how many hours you can study, the same on every day or different for each weekday, and get back a day by day plan with spaced repetition review dates, pomodoro blocks with breaks, and dated milestones for a big assignment. Ask for a revision timetable, an exam schedule or the spaced repetition dates for a topic, and it works them out from your dates and your hours. The date math is done by tested code, not guessed by the model.

Why not just use Brightspace tools

Brightspace and Canvas tools read your school site. They tell you what is due, what your grades are and what your teachers announced, and they need an account and a sign in. This server reads no site at all. You give it the exam dates and the hours you can study, and it works out the timetable, the review dates and the breaks. Run it next to an LMS server if you like: one tells you what is assigned, the other tells you how to study it.

Prompt gallery

Copy and paste any of these into your AI app once the server is installed.

• "Plan 3 weeks of study for my 4 exams"
• "What should I study today?"
• "When should I review what I learned today for my exam on Nov 30?"
• "Plan 3 hours of studying as pomodoro blocks with breaks."
• "Break my 6 hour essay due Friday into milestones."

Ask it things like:

• "I have a networking exam on Dec 12 and scripting on Dec 15. I can study 3 hours a day except Sundays. Make me a plan."
• "I have a networking exam on Dec 12. I study 2 hours on weekdays and 6 hours on weekends. Make me a plan."
• "My networking exam covers routing, switching and security. Split the study hours over those topics and tell me what to study today."
• "When should I review what I learned today so I remember it for my exam on Nov 30?"
• "What should I study today? My networking exam is on Dec 12 and I studied routing last Friday."
• "Am I on track for my exam on Dec 12 if I can study 2 hours a day?"
• "Plan 3 hours of studying as pomodoro blocks with breaks."
• "Break my 6 hour essay due Friday into milestones."
• "Give me this plan as an .ics file so I can import it into Google Calendar."

This is a free, open source MCP server. It works with Claude, ChatGPT, Cursor, VS Code Copilot and any other app that supports MCP. It runs on your computer, so your study data never leaves it. The install steps below have the exact config for each app.

Install

The server runs on your computer through npx, so you need Node.js 20 or newer. There is no account, no key and nothing to sign in to.

Every app reads its MCP config from its own file, so pick yours:

• Claude Desktop
• Claude Code
• ChatGPT
• Cursor
• VS Code Copilot

All of them start the same server the same way. The only difference is where the JSON goes.

Claude Desktop

Add this to claude_desktop_config.json:

{

"mcpServers": { "study-planner": { "command": "npx", "args": ["-y", "study-planner-mcp"] } } }

Restart Claude Desktop after you save the file.

Claude Code

Run this in a terminal:

claude mcp add study-planner -- npx -y study-planner-mcp

ChatGPT

The ChatGPT desktop app can start a server that runs on your machine. Open Settings, turn on developer mode, add a connector and paste this JSON:

{

"mcpServers": { "study-planner": { "command": "npx", "args": ["-y", "study-planner-mcp"] } } }

The ChatGPT website talks to servers over the web, so use the desktop app for a server that runs on your computer.

Cursor

Cursor reads mcp.json from the project folder at .cursor/mcp.json, or from your home folder at ~/.cursor/mcp.json if you want the server in every project. On Windows the home folder file is %USERPROFILE%\.cursor\mcp.json.

{

"mcpServers": { "study-planner": { "type": "stdio", "command": "npx", "args": ["-y", "study-planner-mcp"] } } }

Restart Cursor after you save the file. The Customize page in the sidebar lists the server once it starts, and you can turn it on or off there.

VS Code Copilot

Put this in .vscode/mcp.json in your project folder, or in the mcp.json file in your user profile:

{

"servers": { "study-planner": { "type": "stdio", "command": "npx", "args": ["-y", "study-planner-mcp"] } } }

VS Code also reads a portable .mcp.json file in the project root. That file uses the name mcpServers instead of servers:

{

"mcpServers": { "study-planner": { "command": "npx", "args": ["-y", "study-planner-mcp"] } } }

Restart VS Code after you save the file.

After any of these, try one of the questions at the top of this page.

Tools

The original names make_study_plan, spaced_repetition_dates, pomodoro_plan and assignment_milestones do the same four jobs with the same parameters, so existing configs and saved prompts keep working.

Every date these tools take can be written in plain words, like Dec 12 or in 3 weeks. See Dates in plain words.

Study Planner MCP compared with Brightspace and Canvas

Brightspace MCP and Canvas MCP read your school site. They tell you what is due, what your grades are and what your teachers announced. They need an account and a sign in.

This server reads no site at all. You give it the exam dates and the hours you can study, and it works out the timetable, the review dates and the breaks. Run it next to an LMS server if you like: one tells you what is assigned, the other tells you how to study it.

Brightspace also writes your due dates into a calendar file, and both of them show the dated events your school posted. What they do not do is turn your exam dates and your study hours into a day by day timetable with spaced repetition dates and pomodoro breaks.

Keywords

People search for a study planner with words like these, so here they are in one place. Revision timetable, exam schedule and spaced repetition dates are on the list, along with the other names for the same jobs.

• study planner mcp
• study planner
• study planner offline
• study-planner-mcp
• study plan mcp
• revision timetable
• revision schedule
• revision plan
• revision planner
• reschedule study plan
• exam schedule
• exam timetable
• exam planner
• exam prep
• exam countdown
• exam study plan
• study schedule
• study timetable
• study calendar
• spaced repetition
• spaced repetition dates
• spaced repetition schedule
• pomodoro
• pomodoro planner
• assignment milestones
• assignment plan
• student planner
• student
• mcp
• mcp server
• model context protocol
• offline

The npm listing uses the same words. If you searched for one of them, install the server and ask for a plan.

How the study plan works

Each day has the hours of that weekday, which is hours_per_day unless you set hours_per_weekday. Those hours are shared across the exams that still need work, weighted by how many hours are left divided by how many days remain, so closer and bigger exams get more time. Study stops the day before each exam. If the hours do not fit, the plan says how many are short instead of pretending. The dates the server works out are plain calendar dates, so time zones never shift a day.

When an exam has topics, the hours the exam gets on a day are split over its topics in proportion to the hours each topic still needs. That way the plan answers which topic to study, not only which exam.

Dates in plain words

Every date the tools take can be written the way you say it. The server reads the date itself, so the day math stays in tested code.

• 2026-12-12 for a date you know exactly.
• Dec 12, December 12 or 12 Dec for a month and a day. Without a year it means the next one that has not passed yet, so asked for in October, Dec 12 means 12 December of that year, and asked for in December it means 12 December of the next year.
• Dec 12 2026 when you want a year of your own.
• today, tomorrow and yesterday.
• in 3 days, in 2 weeks, in 3 months and in 1 year, counted from today. Number words work too, like in two weeks.
• 3 days ago and 2 weeks ago for the days behind you.
• Friday for the c