Set a goal with your assistant
Choose a target date, then break the goal into skills and small pieces of knowledge. Your assistant creates questions from your material.
Set a learning goal and a date, or keep learning without a deadline. Your assistant explains; OpenKT tracks what you know and helps choose the next question.
Open workspaceGoals, skill maps, knowledge and learning evidence are saved to your account. Use the same Google account in your assistant and on this website.
Check recall estimates, plan short study sessions, and record what you already know before learning something new.
Choose a target date, then break the goal into skills and small pieces of knowledge. Your assistant creates questions from your material.
Use flashcards, multiple choice or True / False. Quiz answers are checked; flashcard recall is self-reported.
See current recall or a forecast for your target date, with uncertainty and the evidence behind it.
Add this MCP URL in your assistant and sign in with Google. ChatGPT developer mode uses OAuth DCR registration.
https://kt-mcp-hqtzrnf2ma-uc.a.run.app/mcp
Refresh the tool catalog after updates and open a study session in a new conversation. Inline view availability depends on your client.
Install the OpenKT plugin, then run /mcp, choose openkt and sign in with Google. It adds /openkt:learn, /openkt:study, /openkt:review, /openkt:plan and /openkt:paper.
/plugin marketplace add DoniMoon/kt-mcp
/plugin install openkt@openktOpen Settings → Connectors → Add custom connector, paste the MCP URL above and sign in with Google.
Recall estimates change with time and new evidence. Initial knowledge checks establish a baseline; explanations and later assessments are recorded separately. These values can decrease and are not achievement points.
Developer preview with Google sign-in, learning goals, skill maps, flashcards, MCQ/OX and a personal study planner.
Improve the learning model, session experience, and cross-device consistency through real use.
Prepare a distributable ChatGPT plugin after the core product is ready for a wider audience.
PRIMARY CONTRIBUTORPhD student · Human-Computer Interaction Institute, Carnegie Mellon University
ADVISORHuman-Computer Interaction Institute & Language Technologies Institute, Carnegie Mellon University
ADVISORHuman-Computer Interaction Institute, Carnegie Mellon University
Sign in with Google to manage your knowledge, goals and study plan.
Open workspace