#multi-agent-systems

A trilingual (繁中 / English / 简中) learning roadmap for agentic AI: from LLM basics to multi-agent systems, with 240+ curated resources and hands-on examples. 中文 AI agent 學習地圖。

Q-Learning Swarm Foraging 2026: Multi-Agent RL in Dynamic Grid Environments

🔁 Build reliable recurring AI-agent systems: 968 resources, 22 operational patterns, 22 loop contracts, 8 runtime starters, an interactive atlas, and a structured dataset.

Autonomy Composition Abstraction: a machine-readable charter format for defining authority, obligations, constraints, and escalation paths in multi-agent systems

Build and operate production-grade AI agent systems through this 22-chapter technical course designed for collaborative learning with AI partners.

Multi-Agent AI Orchestrator 2026 🚀 | YAML, 6+ LLM Providers, ReAct & Swarm

AgentBridge — an OpenClaw coordination dashboard and API for assigning, tracking, and reviewing work across AI agents.

Tamper-evident audit logging for OpenClaw and other local AI agents: hash-chained, locally verifiable, zero cloud dependency.

An MCP server for orchestrating multi-agent workflows inside herdr, enabling real-time communication, pane management, and contextual handoffs.

Battle-tested multi-agent dev team for Claude Code — 9 specialized sub-agent templates + a 6-stage pipeline with agent memory, verdict protocols, commit checkpoints & standing review prompts. CLAUDE.md constitution, bug-pattern library, full Laravel + React example. Bilingual EN/AR. Extracted from a production app with 200+ releases.

Universal Agent Skill: builds anything with infinite disk memory and a two-brain team — typed consent gates, reproducible claims. Works with Claude Code, Codex, Antigravity.

Agentic loop OS for Claude Code on Windows. State on disk, QA with separate memory, and a rate-limit circuit breaker that pauses at zero tokens and resumes on reset. 47 offline tests.

A multi-agent AI platform leveraging a Master Orchestrator and LLMs to unify and automate corporate finance, business intelligence, sales, and HR workflows.

Governed operating loops and portable capability Capsules for persistent AI agents

Enterprise Multi-Agent Microgrid SCADA & Smart City Traffic Control Cockpit powered by Google Gemini 2.5 AI, 3/3 BFT Consensus, SHA-256 Cryptographic Ledger, Kaggle Machine Learning Predictors, and 10-Country Currency Matrix.

Multi-agent newsroom pipeline: 6-agent hub-and-spoke architecture using Claude Sonnet

Tool-neutral Codex and Claude Code orchestration sample for Sitecore XM Cloud headless development

A privacy-aware multi-agent career management system with structured source governance, human approval workflows, and LLM output evaluation.

Multi-agent AI system that reviews arXiv research papers using LangGraph orchestration and Groq LLMs: reading, critiquing, and analyzing citations through a Streamlit UI where each user brings their own API key.

A protocol for continuous AI operation through rotating active, shadow, and regeneration shifts across distributed agents and model replicas.

Agentic AI portfolio: planning agents, multi-agent orchestration, RAG, memory systems, semantic routing, evaluation pipelines, and business workflow automation.

Synthetic benchmark for role-limited multi-agent deliberation in reservoir construction decision support

Reference implementation of a transport-agnostic protocol for signed, bounded, and verifiable delegation between autonomous agents.

OMEGA INFINITY 1000 — Enterprise AI Orchestration Platform with Agent SDK, multi-agent workflows, deployment automation, and Next.js dashboard

4 AI agents negotiate the same job offer under 3 coordination patterns — orchestrator (24 turns), blackboard (14), hybrid (13). Same deal, different bills. Gamified interactive demo replaying real Claude runs — no API key needed. Teaching guide included.

Multi-agent online store on Azure AI Foundry: shop assistant (13 connected agents) + 6-agent marketing campaign pipeline

Do your agent pipeline's own checks catch planted faults? Measured on LangGraph, CrewAI and AutoGen: median 16.7%. One prompt-level change takes it to 55.0%. Pre-registered spec, Apache-2.0 harness, every raw trace published.

SwarmPIM is a production-grade, containerized PIM system engineered with Next.js 14 and FastAPI. It orchestrates a multi-agent AI swarm utilizing Gemini and Grok with dynamic fallback routing. The architecture features an embedded Qdrant vector cache, high-speed Polars batch ingestion, and analytical DuckDB storage optimized via Apache Arrow.

A staged-gates, multi-agent software delivery methodology for Claude Code — sub-agent definitions, orchestration protocol, quality-gate hooks, and ports for other AI coding tools. From the article 'The Agentic Delivery Office'.

Multi-agent AI backend built on LangGraph — Supervisor routing, Human-in-the-Loop approvals, Gmail & Calendar tools, real-time SSE streaming, and persistent Postgres-backed memory.

Open discovery infrastructure for AI agents. Crawl, validate, normalize, and search A2A-compatible agents.

Production-grade agentic AI on Kubernetes — LangGraph-native agent orchestration (Synapse) on EKS + Istio Ambient Mesh + AgentGateway. Zero-credential LLM access, pluggable MongoDB Atlas Vector Search memory, A2A protocol runtime, multi-tenant by design.

Enterprise multi-agent customer support platform built with CrewAI, Streamlit and OpenAI, featuring direct AI assistance, web-verified responses, case documentation, guardrails, analytics and audit-ready records.

Autonomous multi-agent engineering-ops copilot — LangGraph agents that investigate, fix, and verify bugs using MCP tools, memory, and a bounded retry/human-approval loop. 100% local via Ollama.

Orquestación de IA multiagente: un registro de agentes autogenerado + una evaluación a ciegas que demuestra que el enrutado acierta el agente correcto (27/27, sin regresiones)