A Spring AI library that brings Claude Code-inspired tools and agent skills to your AI applications.

🐔 Agentic systems explained with chickens — the 2024 taxonomy + the 2026 canon (context engineering, skills, MCP, harnesses). Every pattern as a runnable, CI-checked file. Mermaid everywhere.

Hands-on tutorials for building AI agents from scratch. Learn LLM APIs, prompt engineering, tool calling, and the agent loop through practical examples.

Build AI engineering skills from scratch with 260+ lessons, 20 phases, and hands-on tools for practical learning

A simple spring boot app demonstrating how to implement Evaluator Optimizer Workflow pattern using Spring AI

Tech stack: Python, A2A, ADK, MCP, CrewAI, FastAPI, Ollama, Pydantic, Uvicorn, etc.

Reference implementation of Spring AI's Agentic Patterns — a hackable Java/Spring Boot AI agent that runs as a Telegram-native personal assistant with compounding file-based memory, or a one-shot task runner. Same JAR, both modes.

You lead. Agents hunt. Cognitive + methodological harness for AI-assisted software — 139 skills across 7 companion packs + composition layer. Cross-tool (Claude Code, Cursor, Codex, Gemini CLI).

ReflectiveAgent

The CPTO's playbook for agentic AI: patterns, anti-patterns, decision trees, and procurement-grade checklists for shipping reliable agents in production. Vendor-neutral, with 5 runnable examples and an explicit EU AI Act lens.

Building Effective Agents explains the distinction between agents and workflows and introduces common patterns for both. In this project, I implement every workflow and agent pattern described in the LangChain and Anthropic blog from scratch using LangGraph with clean structure, clarity, and reproducible code

Generalizable Orchestrator → Planner → Executor → Reviewer multi-agent pattern using LangGraph. Plug-and-play tool nodes with YAML-configurable agent personas — designed as a reusable blueprint for enterprise ISV agentic AI deployments.

Implementation of a tool-using agent in Python with LangChain. The agent is designed to dynamically select and use tools, such as API wrappers, to answer complex questions that require external data.

An implementation of a tool-using AI assistant for precise calculations. This project, part of a technical assessment, equips an LLM with a calculator to ensure mathematical accuracy.

Implementation of a planning agent using LangGraph's state machines. The agent dynamically creates task lists and executes them sequentially, demonstrating a robust pattern for AI task decomposition.

Claude Code plugin marketplace with reusable agentic skill patterns for reasoning, decision-making, and AI workflows.

Blueprint: Multi-Agent Orchestration Patterns | Published on omnithium.ai

Workflow vs Agent architecture in LangGraph — 6 agentic patterns in one demo (Tool Use, Routing, Parallelization, Voting, Human-in-the-Loop, Orchestrator)

KVKK/GDPR-aware enterprise AI orchestration reference architecture — multi-domain, multi-channel, federated data; conductor pattern.

An implementation of the self-correction and reflection pattern using LangGraph. The agent generates a response, critiques it for errors or omissions, and then re-generates a refined answer.