Report Abuse

Basic Information

This repository provides a combined theoretical overview and practical code samples for building and deploying large language model (LLM) based agents and Model Context Protocol (MCP) tools. It covers LLM architectures, prompt engineering, retrieval-augmented generation (RAG), fine-tuning methods, agent frameworks and protocols such as MCP and A2A. The repo includes hands-on sample projects and agent implementations using AWS Strands and the Google Agent Development Kit, plus multi-agent workflows and containerized examples using FastAPI and Streamlit. It also contains LLM project examples such as an AI content detector and MCP integrations with PraisonAI and Ollama. The material targets developers and architects who want conceptual background alongside runnable examples to experiment with agents, tools, memory, and multi-agent orchestration.

Links

Categorization

App Details

Features
Contains explanatory sections on LLM architecture, prompt engineering, RAG and several fine-tuning approaches including adapters, LoRA and QLoRA. Documents agent frameworks and tools including Google ADK, CrewAI, PraisonAI and PydanticAI. Ships a structured set of agent samples labeled sample-00 through sample-09 demonstrating first agents, containerized agents, local and remote MCP tools, built-in search, memory usage, LiteLLM and AWS Bedrock connections. Provides multi-agent patterns—sequential, parallel, loop and hierarchical—each with Streamlit GUIs. Includes example projects such as an AI content detector and MCP client/server patterns, plus FastAPI-based backends and container examples to illustrate deployment.
Use Cases
This repo helps practitioners learn both theory and practice of LLM-based agents and MCP tooling by pairing concept explanations with runnable examples. Users can follow sample code to build agents that invoke external tools, run local or remote MCP servers, and connect models such as Gemini, Ollama or Bedrock-hosted Llama. Multi-agent patterns and Streamlit user interfaces accelerate prototyping of orchestrated workflows, while FastAPI examples demonstrate containerized front-end/back-end integration. The included projects and curated references support exploration of RAG, persistent memory, fine-tuning strategies and agent-to-agent communication, making it easier to prototype, test and adapt agent architectures for research or production evaluation.

Please fill the required fields*