affaan-m/ECC
ECC is a harness-native operator system designed for agentic AI work, providing a comprehensive solution with skills, memory optimization, security, and research-first development across multiple A...
Awesome AI Agents › Multi-Agent Frameworks
Kocoro-lab/Shannon
Shannon is a platform for running AI agents in production, built around durable workflow execution with explicit cost and safety controls. Tasks enter through a Go gateway that handles the REST API, JWT or API-key authentication and rate limiting, then reach a Go orchestrator that runs Temporal workflows, decomposes tasks, manages budgets and routes by complexity. A Rust agent core acts as the enforcement gateway with a WASI sandbox for code execution, token counting and a circuit breaker, while a Python LLM service abstracts providers, exposes MCP tools and runs the agent loop; a Playwright service handles browser automation and scraping. PostgreSQL, Temporal and Redis back the state. Because executions run as Temporal workflows they can be replayed step by step, which the project calls time-travel debugging, and every run emits real-time event streams, Prometheus metrics and OpenTelemetry traces. Hard token budgets per task and per agent with automatic model fallback address runaway cost, while OPA policies and multi-tenant isolation cover access control, including human approval gates. Provider support spans OpenAI, Anthropic, Google, DeepSeek, xAI, any OpenAI-compatible endpoint and local models through Ollama. Tasks are routed automatically into one of several execution strategies: simple single-agent responses below a complexity threshold, a DAG strategy with fan-out and fan-in dependency tracking as the default for multi-step work, ReAct for iterative reasoning with tools, a research strategy that uses tiered models to cut cost, an exploratory tree-of-thoughts strategy for parallel hypotheses, a browser-use strategy backed by Playwright, domain analysis for specialised deep research, and a swarm strategy in which a lead orchestrates autonomous agents with convergence detection. Clients can call the REST API, stream events over server-sent events, use the Python SDK, point existing OpenAI code at an OpenAI-compatible base URL, or use a Tauri desktop app. A skills system lets users add custom skills in a config directory and select one per task. Installation is a one-command Docker Compose script; MIT licensed.
https://github.com/Kocoro-lab/Shannon
Also in Multi-Agent Frameworks
ruvnet/ruflo
Ruflo is a leading agent orchestration platform for Claude, enabling the deployment of intelligent multi-agent swarms, coordination of autonomous workflows, and building conversational AI systems.
Yeachan-Heo/oh-my-claude-sisyphus
Oh My Claude Sisyphus (OMC) is a multi-agent orchestration framework for Claude Code, simplifying complex AI agent workflows without a steep learning curve.
microsoft/JARVIS
JARVIS is a collaborative AI system that connects large language models with expert AI models from HuggingFace to autonomously plan, execute, and integrate complex AI tasks.
rowboatlabs/rowboat
Rowboat is an AI-powered platform that enables users to build and manage multi-agent swarms using natural language, integrating tools and knowledge for automated workflows and deployment via API or SDK.
cft0808/edict
Edict is an OpenClaw Multi-Agent Orchestration System featuring 9 specialized AI agents, a real-time dashboard, model configuration, and full audit trails, inspired by ancient Chinese imperial gove...
MemoriLabs/Memori
Memori is a versatile SQL native memory layer designed to enhance LLMs, AI agents, and multi-agent systems with advanced memory management, seamless integration, and high performance.
HKUDS/DeepCode
DeepCode is an open-source project that uses multi-agent AI systems to convert research papers and natural language into production-ready code through both CLI and web interfaces.