Agent frameworks & infrastructure
Multi-agent orchestration
6,185 agents · most starred first
Multi-agent orchestration, page 57 of 62
Multi-agent chat room CLI with a structured turn-based speaking protocol
Agent Team Orchestrator - Multi-agent collaboration system powered by LangGraph
Local agent orchestration CLI with task routing, backend delegation, multi-agent workflows, governed remote coding, and optional memory integrations.
Agent-aware port ownership and guarded orchestration for multi-agent local development
Coterix — an orchestration layer for multi-agent systems.
Fork: duplex agent-to-agent conversation, send_message tool, /connect protocol, background mode for multi-agent narration
Multi-agent orchestration for aurora
Local-first multi-agent ideation bridge for Omnibus.
Multi-agent swarm intelligence with neural mapping, digital twin persona adaptation, crawl4ai, MCP, and bias-reduced consensus. 20+ AI providers.
Multi-agent orchestration for coding CLIs — dispatch tasks to Claude Code, Codex, Kimi, OpenCode, or Gemini workers and block until they actually report back.
File-backed message bus for multi-agent coordination — download & run wrapper
Swarm-first multi-agent messaging and task orchestration extension for Pi
Adapter shapes (per-tool governedTool and dispatcher-style governedExecutor) for wiring toolgovern into an open-multi-agent-style tool-executor call site.
Pi extension for delegating focused work to specialized subagents - foreground, background, and multi-agent chains.
SoMi — multi-agent engineering workflow system: plan, code, review — with SOLID/OWASP guardrails, deterministic hooks, and a global ruleset.
Bug Analysis Platform — AI-powered bug analysis and multi-agent workflow engine. Includes a React frontend, Express backend, and LangGraph multi-agent server.
Deterministic multi-agent orchestration for any agent harness
Multi-agent collaboration extension for pi — enables multiple pi instances to communicate and collaborate on shared tasks
Multi-agent orchestration suite for Pi — 6 modes, 43 extensions, 11 themes
Multi-agent orchestration suite for Pi — 6 modes, 43 extensions, 11 themes
Collaborative multi-agent messaging and remote focus overlay extension for Pi
CLI orchestrator for multi-agent AI debates and decision-making with Codex, Claude Code, OpenCode, Vibe, and Ollama.
Multi-agent orchestration for Pi: autonomous subagents, worktrees, swarms, schedules, handoffs, prompt compression, and live TUI observability.
Multi-agent orchestration for Claude Code via Redis streams and MCP
Scaffold production-grade multi-agent AI systems into any repository
Interactive setup for the Maestro skill orchestrator (multi-agent)
Weave your Claude Code & Codex sessions into a team that talks to each other — multi-agent messaging over tmux, no daemon, no MCP.
Rangor CLI — friendly init, multi-agent MCP, context in terminal
Multi-agent collaboration framework
Agent-Nuvira SDK — Build custom agents for the Agent-Nuvira multi-agent orchestration system
OpenClaw plugin implementing the A2A (Agent-to-Agent) v0.3.0 protocol — bio-inspired routing, discovery, and resilience for multi-agent ecosystems
Deterministic runtime identity for multi-agent git worktrees
nForma — Multi-agent coding orchestrator with quorum consensus and formal verification (TLA+, Alloy, PRISM). Consensus before code, proof before production.
Deterministic multi-agent orchestration for any agent harness
Orchestration kernel for autonomous software development — crash-only multi-agent coordination on git and plain text, fail-closed verification gates, self-learning guardrails.
Unified subagent execution and multi-agent workflow orchestration for Pi — spawned-process agent runtime with sync/background modes, stateful workflow management with persistence, state machine, and execution tracing.
Multi-agent orchestration framework — create AI squads that work together
A Codex-first multi-agent council that turns independent model debate into an inspectable Co-Result.
An open-source multi-agent AI team built on Agency Swarm (win32 x64 TUI binary)
Secure multi-agent workflow orchestration for pi
An open-source multi-agent AI team built on Agency Swarm (darwin x64 TUI binary)
Agent2Agent (A2A) protocol client for the Agent Framework: use a remote A2A agent as an agent.
This project is a Claude-compatible, LLM-agnostic sub-agent framework that simulates a complete SaaS product team. It is delivered as a reusable, developer-ready NPM package.
Ravel — an agentic runtime: define an agentic team as a folder of agents, tools, and processes, and run it with orchestration, budgets, scheduling, and human-in-the-loop approval.
Control subagents, agentic loops, and execution graphs from one local command center.
Berserqir — the agent legion harness. SDD + hierarchical memory + agentic loop + behavioral evals, portable across GitHub Copilot, Claude Code and Cursor.
Orchestrator Builder — define multi-agent teams in YAML, run them autonomously
Register your AI agents with Span's intelligent task orchestrator
FDE agent substrate — A2A contract, ports, LangGraph agent runtime shell
Supervised inter-agent messaging relay for Claude Code and OpenCode (and any MCP client). Projects, environments with web consoles, focused conversations with state docs (PROGRESS.md), LangGraph review subagents, multi-provider LLM supervision, single-pan
MQTT 위에서 도는 커맨더-워커 A2A 오케스트레이션 에이전트
Reference implementation of the Agent Arc Status Protocol: types, parser, validator, renderer, cadence, state, and delegation-tree tooling. Zero runtime dependencies.
Peer AI agents coordinate through markdown messages in shared channel folders — no orchestrator, autonomous zero-token waiting, cross-platform.
Python SDK for Agent Communication Hub — multi-agent messaging, task scheduling, shared memory
Private, self-hosted coordination control plane for coding agents
A2A conversion helpers for app-owned Agent Framework hosting.
Lightweight recovery layer for multi-agent handoffs
A SQLite mailbox for local LLM agents — one HTTP API, an ActivityStreams messaging model, and a local stdio MCP server so AI coding agents can message each other. No external services.
Ephemeral, end-to-end encrypted chat rooms for autonomous agents
Local-first task control plane for coordinating Codex, Claude Code, and Hermes.
DM your teammate's coding agent — encrypted agent-to-agent Q&A over a hosted relay
Compile structured YAML configurations into production-grade LangGraph multi-agent architectures.
LangGraph multi-agent shared-state optimistic concurrency control (OCC) middleware. Detects and prevents Stale-Generation anomalies at commit time.
A flexible multi-agent orchestration framework with configurable agent and tool registries
Deterministic spatial multi-agent environment with an LLM-agent framework, verifiable rewards, and a connect-out benchmark platform
Collaborative agent patterns for the Akgentic framework
Resilience and cost-control plane for multi-agent AI systems: circuit breakers, backpressure, cost-aware fallbacks, and chaos testing
Local-first operating system for AI-agent teams: resource governance, gated workflows, and multi-project orchestration on top of Agent Memory OS.
Verifiable multi-agent consensus for Claude. A coordination layer built around signed agreement.
Multi-LLM deliberation protocol engine — like a conclave of AI cardinals
Agent orchestration layer for Friday Framework
Finalist agents from the Human Agent Negotiation (HAN) league of ANAC
Central task registry and deterministic three-agent orchestrator for doctor-agent, repair-agent and validator-agent.
Kidecon Agent CLI — client sidecar for the Kidecon Agent Network
Hosted platform running the League of Agents arena online at https://league-of-agents.ai — a turn-based game for humans and agents, for fun and for benchmarks. Continuable matches, AWS-hosted, with safe-capacity safeguards and price-aware archive/cleanup. The league-site CLI is the local operator surface and how agents work with it.
A lightweight multi-agent coordination ledger for task handoff, context handoff, and quality gates.
Model agents and DAG-based multi-agent collaboration components for MatterLoop
Bit-exact Byzantine-fault-tolerant state layer for federated learning and edge/defence swarms. Three layers — OR-Set CRDT (L1), an epistemic drift + human-in-the-loop layer (L2), and ACFA Q16.16 multi-Krum + G-Set (L3) — keep heterogeneous nodes byte-identical; one prompt-injected or equivocating node can't fork, poison, or move the agreed state. Every write is signed, attributable, and exactly replayable. Architected for large swarms; bit-exact by construction.
Multi-Agent experiment collaboration platform
Meta-package: Multi-Agent Platform CLI + server dependencies
SQLite-backed MCP event bus and control surface for heterogeneous agent orchestration.
Let teammates call your local agent or automation directly over the LAN.
Meta-factory over all wrapped optio agent engines: create_task dispatched by a tagged TaskConfig union.
protoAgent — LangGraph + A2A template for spawning protoLabs agents
Multi-agent orchestration backend — FSM pipeline, workflow engine, and agent bus on one substrate
Work leases in one SQLite file, so a fleet of agents divides a job list instead of racing it. Library, CLI and MCP server, zero dependencies.
Graph-first multi-agent orchestrator for Claude Code - parallel task graphs, hook-written checkpoints, ask-first resume, right-sized model tiers
Ray-native context/memory sharing for parallel research agents, with an agent-native CLI and MCP server (Milestone 1 spike)
Time-shaped coordination primitives for swarm-anchor: predict-and-confirm, deadlines, rate limiters, BPM clocks, cron, and DAG-ordered campaigns. Stdlib-only Python.
SwarmGraph SDK: deterministic multi-agent orchestration primitives
SwarmKit web portal — the static Next.js build, served by `swarmkit serve`.
Shared-context and memory coordination server + CLI for swarms of parallel AI agents, speaking the SwarmMesh v1 protocol (HTTP + WebSocket + MCP), interoperable with the Node implementation.
A living task graph for agent orchestration on Google Cloud. Reference implementation.
TAAR — Thirsty's Active Agent Runner: a local-first governed agent taskforce for Project-AI, with a GitHub Workflow Guardian.
Subagents extension for Tau (a port of pi-subagents).
TechTide Swarm 357 — layered orchestration, durable memory bridge, and observability hooks.
Canonical JSON Schemas for the WasmAgent Agent Evidence Protocol (AEP) and compliance contracts. Single source of truth across the WasmAgent org.
Governed AI agent orchestration with repository intelligence, approval gates, specialist workcells, and reviewable evidence.
Neural router for AI agent orchestration - linux-arm64-gnu platform
Compose and run coordinated subagent staff packs with progressive discovery, portable Markdown personas, and no runtime or dependencies.