- Home
- Skills
- Agents & Orchestration
- AI-Tutor
AI-Tutor
Works with the AI Agents you already use
AI-Tutor
AI tutor agent using LangGraph for RAG-based PDF tutoring and human-in-the-loop video summarization.
$1
AI-Tutor
Example session with this skill installed
Upload the file 'quantum_mechanics_intro.pdf' and explain the Schrodinger equation based on the text.
- Read your context and instructions
- Compiled the ai-tutor
I have processed 'quantum_mechanics_intro.pdf' and stored the embeddings in the FAISS vector store. Here is the explanation grounded in your document.
Explanation: The Schrödinger Equation
Based on the provided text, the Schrödinger equation is the fundamental law of motion for quantum systems. It describes how the quantum state of a physical system changes with time.
| Concept | Description | Mathematical Form |
|---|---|---|
| Time-Dependent | Describes the evolution of the wave function over time. | $$i\hbar \frac{\partial}{\partial t} \Psi(\mathbf{r},t) = \hat{H} \Psi(\mathbf{r},t)$$ |
| Hamiltonian ($\hat{H}$) | The operator representing the total energy (kinetic + potential). | $$\hat{H} = -\frac{\hbar^2}{2m}\nabla^2 + V(\mathbf{r},t)$$ |
| Wave Function ($\Psi$) | A complex-valued probability amplitude for the particle's position. | N/A |
Key Takeaways from the PDF:
- The equation is deterministic, even though the interpretation of the wave function is probabilistic.
- Energy quantization arises naturally from the boundary conditions applied to this equation.
Next steps
- Ask for a specific example of the "Infinite Square Well" calculation mentioned in chapter 2.
- Upload a lecture video URL to compare the author's explanation with a verbal lecture.
- Provide feedback to simplify the mathematical notation for an introductory level.
Connects securely to your tools. The creator never sees your data.
About this skill
Students struggle to synthesize information from dense textbooks and long lecture videos. This skill delivers an agentic tutor that grounds its reasoning in your specific PDFs and video content to provide accurate, multi-modal instruction.
What it does
- PDF-Grounded RAG: Ingests textbooks or notes into a FAISS vector store to provide context-aware answers that prevent hallucinations.
- Video Transcription: Uses Whisper to locally transcribe audio from
.mp4files or YouTube URLs when native transcripts are missing. - Iterative Summarization: Generates summaries from video transcripts and revises them based on human-in-the-loop feedback.
- LaTeX Math Rendering: Formats all scientific formulas and equations using clean Markdown math blocks for readability.
- Multilingual Tutoring: Supports instruction and retrieval in both English and Arabic using multilingual embeddings.
How it works
- Knowledge Ingestion: Upload a PDF or provide a video source (local path or YouTube link) to initialize the knowledge base.
- Agentic Retrieval: Submit a question. The LangGraph agent decides whether to answer from the LLM or retrieve specific chunks from the FAISS index.
- Summary Review: For video tasks, review the initial AI-generated summary and provide specific refinement instructions.
- Final Approval: Iterate on the output until satisfied, then type
saveto finalize the tutor's summary.
Frameworks & tools
Built on Python 3.10 and the LangChain ecosystem, including LangGraph for state management and LangSmith for monitoring. It uses Google Gemini for reasoning, FAISS for vector storage, Whisper for local transcription, and Gradio for the interface.
Why this beats prompting it yourself
Manual RAG setups often fail at complex reasoning or multi-modal integration. This skill uses structured LangGraph workflows with human-in-the-loop nodes, ensuring the agent doesn't just guess but actually verifies its answers against your specific study materials.
Use cases
- Summarize a two hour university lecture video into five key study points.
- Query a specific textbook PDF for formulas and step-by-step math derivations.
- Refine automated transcripts into concise, bulleted exam revision notes.
- Bridge language gaps by tutoring in Arabic using English language source materials.
Known limitations
Requires a Google Gemini API key. Local video transcription performance depends on host hardware for the Whisper model.
Requires Google API key for Gemini. Local transcription speed depends on hardware. YouTube downloads require yt-dlp.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
Requires