LangChain4j + Quarkus playground (RAG, MCP, A2A, ...)

modelcontextprotocol java sdk

🤖 Build personalized intelligent agents easily with AgentX, a platform that simplifies creation using natural language and integrated tools.

[DOC] Some experimentations with ChatGPT and IGNF APIs (french)

LLM engine with MCP capabilities and Discord interface

My MCP servers

A TypeScript-driven backend project designed to power a rich, customizable portfolio experience, and show-case of others Projects

Python-based Telegram notification service with MCP (Model Context Protocol) support for AI assistants. Built with aiogram v3.0+ for async operations, featuring comprehensive error handling and testing.

Lightweight, educational cross-language protocol explorations with SSE and stdio transports, plus a minimal embedded Java example.

All-in-one web analysis toolkit with JS-rendered HTML, content extraction, visuals, and audits tailored for developer workstreams.

A TypeScript MCP playground showcasing host and server interactions, tool calls, and function calling patterns for rapid experimentation and learning.

AI-powered culinary assistant delivering ingredient-based recipe generation, personalized guidance, nutrition insights, semantic search, and automated validation for user preferences.

Multi-service backend coordinating AI workflows via a unified gateway with JWT enrichment and Kubernetes-ready deployment.

Modern, full-stack Kanban with FastAPI/Next.js, real-time updates, clean architecture, and MCP-powered AI tool calls.

TypeScript toolkit for building and monitoring context-driven clients and services with strong typing, observability, and streamlined integration.

Node.js MCP-style integration for Noko time-tracking with rich filters, date helpers, and safe, confirm-based edits.

Spring-based context propagation with OpenAPI contracts, Kubernetes deployment, and observability-first design for scalable distributed apps.

A Python MCP playground with multiple backends and an inspector workflow for rapid prototyping and learning.

AI-powered testing for Flask apps with Playwright integration and automated test generation from page analysis.

R-based extension of the MCP-Mod framework for binary endpoints enabling Phase 3 success simulations.

Rust-based server-client enabling AI-driven desktop automation on macOS, including app launches, screen reading, clicks, and input.

Low-latency, voice-driven AI for macOS that streams audio to OpenAI, with a local bridge and keyboard-approved tool calls for hands-free automation.

Lightweight Python utilities for Global Fund analyses with mkdocs-based documentation and practical setup guidance.

Bridge between AI models and PromptHub with blockchain-backed prompts, DAGs, and flexible access control.