shiping
An interview assistant and an exam assistant that uses AI to provide solution ideas in real time during 「面试」 or 「在线考试」.
AI translation
Photograph a maths question. Hint gives one guiding question, or Solve gives the answer and the steps, inside the selected course from Algebra 1 through Multivariable Calculus.
EducationTypeScript · CSSGitHub @adhootOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Solve math questions from a screenshot
Translation pending
Taken from the repository README
Loading the full README.
Read the full README ↗Photograph a maths question from a textbook, worksheet, or screen. Pick a course, then choose Hint or Solve. Hint follows the tutor prompt for that course: one short hint or guiding question, and not the final answer. Solve gives the answer and the steps, using only ideas taught up through that course. The courses, in order, are Algebra 1, Algebra 2, Geometry, PreCalc, Calculus AB, Calculus BC, and Multivariable Calculus. Earlier courses get shorter steps and simpler words. If a problem needs a later course, the reply says so and stops at the selected course. Answers, steps, hints, and follow-ups render Markdown and LaTeX.
npm install
npm run dev
The dev server listens on http://127.0.0.1:43123 (0.0.0.0:43123).
Take a picture opens this device's camera in the page and captures a frame. Upload a photo picks an image you already have. Hint and Solve are the two run buttons. You can select one before the photo, and you press it again after the questions are listed. Crop the photo before anything else happens. Math Buddy then reads every question in the crop and waits. Choose all of the questions, or pick one from the list, then press Hint or Solve. The selected course is filled into the tutor prompt for a hint, and it sets the course ceiling for a solution.
Each finished hint or solve is saved in this browser with IndexedDB. The record keeps the course, the mode, the cropped picture, the questions that were found, which of them were used, the hint or the solution, the follow-up thread, and the model calls for that session. A model call is the question read, the hint or solve, or a follow-up: the course, the mode, the start and finish times, the text request, and the raw model reply. Those calls stay on the device. History opens a saved session on the page. The model calls for that session appear only in History, after the session is opened. A follow-up uses the same mode as that session.
The route POST /api/solve reads the first key it finds:
| Order | Environment variable | Model |
|---|---|---|
| 1 | OPENAI_API_KEY | gpt-4.1-mini |
| 2 | XAI_API_KEY | grok-4.6 |
| 3 | GROQ_API_KEY | qwen/qwen3.8-27b |
| 4 | ANTHROPIC_API_KEY | claude-sonnet-5 |
Copy .env.example to .env.local and set one key. Restart the dev server after you add it. If none of these variables is set, the page still loads and the solve button returns a clear message that a key is missing. Do not invent a key.