M

multi-agent-systems

Power Transparency Network (PTN) is an AGPL 3.0+ modular intelligence system that analyzes how energy is produced, controlled, owned, and experienced across the United States. It maps state-level energy generation, vendor concentration, and ownership sovereignty while extending into household-level energy independence modeling. The system is designed to expose structural dependencies in energy markets and provide transparent, data-driven insight into monopoly risk, energy resilience, and domestic infrastructure control. https://roxanneardary.com/power-transparency-network/

CNC Companion is an open-source AI tutor for apprentice machinists. It provides step-by-step lessons, real-time guidance, skill tracking, and immersive learning tools for CNC and other shop floor machines. With adaptive instruction, AR support, and a community-driven lesson marketplace, CNC Companion helps novices learn safely, improve precision, and master machining from beginner to expert. https://roxanneardary.com/cnc-companion/

Yet another software engineering agent orchestration system. This time with a mythical theme. Originally designed for Claude Code, now with support for OpenCode, Github Copilot, and OpenAI Codex.

Contains the fundamental classes and interfaces that define the basic functionality of the SCOP framework.

Commons Reasoning Network is an open-source, AGPL 3.0+ federated intelligence infrastructure designed to support autonomous agents, distributed compute, and interoperable AI systems. It provides a modular platform for reasoning, memory, identity, and economic coordination through an open protocol (UAP), enabling scalable agent ecosystems, cooperative governance, and public-interest AI infrastructure without centralized control. https://roxanneardary.com/commons-reasoning-network/

LatticeOS is a modular, human-in-the-loop operating system for governed multi-agent AI orchestration, designed to provide structured execution, deterministic workflows, and full auditability. Built around event-sourced transparency and explicit governance controls, LatticeOS enables intelligent systems where every agent action becomes structured, traceable history while keeping humans in control of all critical decision-making.

A production-grade, multi-agent urban intelligence framework. Orchestrates NYC 311 data triage and autonomous remediation using LangGraph and a distributed Exo-sharded inference cluster.

Autonomous AI agents that secure, test, and deploy code through intelligent DevSecOps workflows. Built for GitLab Duo Agent Platform to automate PR review, risk detection, compliance checks, and deployment decisions.

distributed cognitive framework in Elixir/OTP.

It combines Nabla-Infinity recursive introspection (∇∞), blackboard-based shared memory, and multi-agent systems with LLM integration.

Designed for research, reasoning, training simulations, and advanced AI applications.

[2025] Autonomous Claude Code plugin with pattern learning and skill auto-selection. Features 27 specialized agents across 4 collaborative groups, quality control automation, and intelligent task execution without human approval at each step.

We created a new layer for the popular JADE agent-based platform to support both the creation of a fully decentralized peer-to-peer network of JADE platforms and the execution of services within that network, thus enabling JADE multiagent systems (MASs) to behave as peer-to-peer networks. The layer is created and managed by a set of new JADE agents, which can be added to any JADE MAS. The project has been described in this paper: https://onlinelibrary.wiley.com/doi/full/10.1002/spe.2657

Multi Agent Hierarchical Actor Critic (docu: https://vg_tum.gitlab.io/mahac_rl)

Rust crate to translate between 1D indices and 2D coordinates with wrapping https://crates.io/crates/wrapping_coords2d

Reliable and efficient agent-based models in Rust

A very simple multi-agent system to modelize particles collisions in a 2D discrete environment.