Loading repository data…
emredeveloper / repository
This is a repository of agent notebooks that contains studies on how to use various agent libraries with different tools and methods.
A transparent discovery signal based on current public GitHub metadata.
This score does not audit code, security, maintainers, documentation quality, or suitability. Verify the repository and its current documentation before adoption.
This repository showcases multiple agent systems and LangGraph workflow examples built with modern AI tooling. All projects are fully documented in English with comprehensive examples and usage instructions.
The most advanced project: extracts transcripts from YouTube videos and enables smart question-answering with a modern UI.
✨ Features:
🚀 Quickstart:
cd "Youtube Video - RAG - Agent"
streamlit run streamlit_app.py
A comprehensive multi-agent workflow for CSV data analysis using Gemini Code Execution. Executes real Python code for statistical analysis, visualization, and anomaly detection.
✨ Features:
🚀 Quickstart:
cd "Sequential Agent"
python langchain_seq.py
Configuration:
Set your Gemini API key in langchain_seq.py:
Selected from shared topics, language and repository description—not editorial ratings.
ZhijingEu /
This is a Jupyter Notebook which explains how to use LangChain and the Open AI API to create a PandasDataFrame Agent. This notebook is accompanied a more detailed Medium article https://zhijingeu.medium.com/building-a-natural-language-enabled-pandas-dataframe-agent-with-langchain-openai-f23ac12180df#50a5
31/100 healthfedericozanini /
This work is about creating a DQN agent, able to learn and well perform in the Atari2600 Pong, using Reinfrocement Learning and Deep Q-Network. The code is developed on a jupyter notebook in python while the game environment is provided by OpenAi called "OpenAi Gym".
32/100 healthGEMINI_API_KEY = "your_api_key_here"
Workflow:
Provides a simple multi-agent flow (MathAgent, WriterAgent) with an orchestrator, powered by LM Studio's OpenAI-compatible server.
✨ Features:
🚀 Quickstart:
cd A2A-Agent
# Run in separate terminals
python math_agent.py
python writer_agent.py
python orchestrator.py
Examples built with the LangGraph library for building stateful, multi-actor applications.
langraph_basic.py)Basic loop: user message → LLM → repeat (stops if response contains "done")
flowchart LR
U[Message] --> LLM[llm_node]
LLM --> C{is "done" included?}
C -->|No| LLM
C -->|Yes / MAX_TURN| E[End]
langraph_stream_memory.py)Thread-based memory with InMemorySaver (thread_id isolates conversation history)
flowchart TB
subgraph T1[Thread 1]
Name[Step Will] --> G1[Graph]
G1 --> M1[(Memory)]
M1 --> A1[Answer 1]
A1 --> Recall[Do you remember the step?]
Recall --> G1
end
subgraph T2[Thread 2]
Recall2[Do you remember the step?] --> G2[Graph]
G2 --> M2[(Memory)]
M2 --> A2[Answer 2]
end
langraph_branch_personas.py)Run the same prompt across different personas, then compare results (diff modes)
flowchart LR
P[Prompt] --> F1[Warm persona]
P --> F2[Formal persona]
P --> F3[Instructor persona]
P --> F4[Skeptical persona]
F1 --> R1[Answer 1]
F2 --> R2[Answer 2]
F3 --> R3[Answer 3]
F4 --> R4[Answer 4]
R1 --> COL[Summary Table]
R2 --> COL
R3 --> COL
R4 --> COL
COL --> DIFF[Diff Analysis]
langraph_dynamic_temperature.py)Classify prompt type and select temperature automatically; optional comparison vs fixed temp
flowchart LR
P2[Prompt] --> CLS[Heuristic Classification]
CLS --> DYN[LLM dynamic]
P2 --> FIX[LLM fixed]
DYN --> CMP[Comparison]
FIX --> CMP
🚀 Quickstart:
cd Langraph
# Set environment variables
set LG_BASE_URL=http://127.0.0.1:1234/v1
set LG_API_KEY=lm-studio
set LG_MODEL=google/gemma-3n-e4b
# Run examples
python langraph_basic.py
python langraph_stream_memory.py
python langraph_branch_personas.py --prompt "Write a short motivational sentence"
python langraph_dynamic_temperature.py --prompt "Translate to French" --compare
Features:
Educational project demonstrating tool calling with Google's Gemini AI. Shows both manual (educational) and production (recommended) approaches.
✨ Features:
google_search - Web search using DuckDuckGoscrape_url - Web scraping with BeautifulSoupget_current_weather - Weather data from Open-Meteo APIcalculate_math - Safe mathematical expression evaluationget_current_time - Real time for any timezonewikipedia_search - Wikipedia article summariesget_exchange_rate - Real-time currency exchange rates🚀 Quickstart:
cd "Tool Calling From Scratch"
# Set API key
set GEMINI_API_KEY=your_api_key_here
# Run application
python app.py
Menu Options:
Advanced agent system using Groq API with rate limit management, ReAct agent pattern, and web search capabilities.
✨ Features:
🚀 Quickstart:
cd "Groq - Mixture of Agents"
# Set API key
set GROQ_API_KEY=your_api_key_here
# Run agent
python advanced_agents.py
Files:
advanced_agents.py - Main agent implementationduckduckgo_agent.py - Web search agentapp.ipynb - Jupyter notebook examplesIntelligent agent that lets you interact with MongoDB databases in natural language. Supports dynamic collection detection and automatic schema analysis.
✨ Features:
🚀 Quickstart:
cd "Mongodb SQL Talk"
# Start MongoDB and LM Studio first
python mongodb-langchain-agent-clean.py
Open http://localhost:5000 in your browser.
Example Queries:
Web search integration with Ollama local LLM for enhanced agent capabilities.
✨ Features:
🚀 Quickstart:
cd Ollama
# Start Ollama first
ollama serve
# Run agent
python web_search.py
Collection of advanced agent projects including RAG agents, SQLite storage, structured output, and Ollama integration.
✨ Features:
🚀 Quickstart:
cd Agno
# Install dependencies
pip install -r requirements_rag.txt # For RAG agent
pip install -r requirements_advanced.txt # For advanced features
# Run specific agent
python ollama-rag-agent.py
python csv_analysis.py
python Structured-output.py
Files:
ollama-rag-agent.py - RAG agent with Ollamacsv_analysis.py - CSV data analysissqlite-storage.py - SQLite storage integrationStructured-output.py - Structured output generationapp.py - Main applicationPython execution agent using Phidata framework for code execution and agent management.
✨ Features:
🚀 Quickstart:
cd Phidata-Agent
python python-execute-agent.py
Agent framework example using AgentScope for multi-agent systems.
✨ Features:
🚀 Quickstart:
cd AgentScope
python agentscope_example.py
FastAPI-based agent framework with web interface for building agent applications.
✨ Features:
🚀 Quickstart:
cd "BeeAI Framework"
# Run FastAPI app
python fastapi_app.py
# Or run Flask app
python app.py
Open http://localhost:8000 (FastAPI) or http://localhost:5000 (Flask) in your browser.
General AI agent system with customizable agent configurations.
✨ Features:
🚀 Quickstart:
cd General
pip install -r requirements.txt
pyt
Vishnu345 /
A Chat-bot is a software application used to conduct an online chat conversation in the form of text or speech, instead of providing direct contact with a live human agent. Designed to convincingly simulate the way a human would behave as a conversational partner. With the chatbot, we don’t want to wait for customer helpline, they can easily clarity their doubts within seconds. A chatbot is used in many areas like WhatsApp, messenger, customer support, food order, personal finance assistance, Telegram, search and track flights, and many more. The main objective of this paper is to develop a Chatbot using Python in Jupyter notebook software and to develop an GUI interface using Tkinter library and trained Chatbot with text files containing information about some sample college to help the students to know about it. In this paper Chatbot will give response both in the form of text and Speech.
29/100 healthdearnidhi /
This repository explores **Retrieval-Augmented Generation (RAG)** in depth, covering both foundational concepts and advanced agentic architectures. The project is organized as a **progressive series of Jupyter notebooks**, each focusing on a specific RAG component, from query transformation to adaptive agents and knowledge graph integration.
35/100 healthThis project implements a reinforcement learning (RL) agent that learns to play the classic Tic-Tac-Toe game. It uses a simple value-based approach to learn optimal moves through self-play. The project is implemented in Python using Jupyter Notebook and provides a great foundation for understanding basic RL concepts.
44/100 healthchicolinux /
These Jupyter Notebooks and Python scripts represent my journey into Agentic AI. I hope this work is also helpful to you.
37/100 health