Your AI study partner — not your shortcut.
Build an agent that turns your notes into quizzes, coaches your essays against the rubric, and finds real cited sources — with integrity guardrails that never do your graded work for you.
The integrity contract
AI that makes you smarter, not AI that gets you flagged. These four rules are baked into every EDU template's instructions.
Four study partners, ready to build
One click opens the Builder with the full brief — guardrails included — already written for you.
From zero to study agent in three steps
Pick a template
Choose a study partner below — or describe your own in plain English. No coding, no setup.
Get a guardrailed blueprint
The Builder generates a full agent blueprint — tools, model, and the integrity rules baked into its instructions.
Study with it
Test it live in the sandbox on the free trial, then keep it running with your own API key or a paid plan.
Priced for a student budget
Start on the free trial — 1,000 credits, no card required — and every template above works on it. Bring your own API key (Anthropic, OpenAI, or Moonshot/Kimi) and it's always free with zero markup. Need hosted, always-on agents? Paid plans start at $14.99/mo — and the exact cost is shown before every run.
See honest pricingQuestions students actually ask
Will these agents write my essay or do my homework?
No — by design. Every EDU template ships with an academic-integrity guardrail encoded into the agent's instructions: it coaches, quizzes, outlines, and critiques, but refuses to produce work you'd submit as your own. That's the point — you learn the material instead of outsourcing it.
How much does it cost for a student?
You can start free: the trial includes 1,000 credits (1 credit = $0.001 of real compute), enough to build and test your first study agents. Bringing your own API key is always free with zero markup — you pay the model provider directly. Paid plans start at $14.99/mo if you want hosted, always-on agents.
Do I need to know how to code?
No. Pick a template and the Builder generates the complete agent — instructions, tools, model choice, guardrails — from plain English. You can test it in the sandbox immediately. Developers can export the TypeScript and extend it.
Which AI models can my study agent use?
Frontier models only — Claude Opus for blueprints, GPT-5.5 for sandbox chats, and you can bring your own key for Anthropic, OpenAI, or Moonshot (Kimi). Kimi's long context window is a great fit for feeding a whole semester of notes to a Study Buddy.
Can I use it for high school, college, or grad school?
Yes — the templates adapt to your level because you feed them your own materials: AP notes, college lecture slides, or graduate research papers. The guardrails stay the same at every level.
Is using an AI study agent cheating?
Used the way these templates are built — quizzing you, explaining concepts, coaching drafts, finding cited sources — no more than a tutor or study group is. What crosses the line is submitting AI-written work as your own, which is exactly what our guardrails refuse to do. Always follow your school's specific AI policy.
Learn it. Don't fake it.
Build the study partner that makes you unbeatable on exam day — and keeps your academic record clean.