multi-agent-systems
12 repositories
A meta-harness evolving Claude Flow's multi-agent swarm tooling — Ruflo is an agent meta-harness for Claude Code and Codex, tracing its lineage to the original Claude Flow project. It adds adaptive memory, self-learning behavior, federation, and vector RAG support, installable as Claude Code plugins for developers building multi-agent swarms and conversational AI systems.
A multi-agent AI tutor that builds a lifelong personalized learning memory — DeepTutor is an open-source AI tutoring system from HKUDS using multi-agent RAG and deep research to adapt to each learner over time. It targets interactive learning rather than one-off Q&A, with a CLI tool and a clawbot component for hands-on study sessions. Useful for developers building personalized education or research-agent tools.
Multi-agent orchestration that wraps Claude Code with zero learning curve — OMC adds multi-agent orchestration on top of Claude Code, aiming for teams rather than solo tinkering, with no new workflow to learn. The project also spawns sibling tools for Codex (oh-my-codex) and a lighter SDK-based option (gajae-code), suggesting a growing family of agent-runtime orchestrators for teams scaling parallel AI coding workflows.
Persistent file-based planning for AI coding agents and long-running tasks. Crash-proof markdown plans, session recovery after /clear and compaction, per-turn re-injection against context rot, deterministic completion gate. Manus-style. Install from npm, the Claude Code plugin marketplace, or npx skills. Codex, Cursor, OpenCode, 60+ agents.
An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation
An open-source SDK for building and controlling AI agent harnesses — Strands Agents' harness-sdk lets you build an agent harness in Python or TypeScript and control it end-to-end, model- and cloud-agnostic. It ships as separate Python/TypeScript harness packages plus a CLI, aimed at developers building production agents rather than quick demos.
Turning scattered AI coding agents into a managed team — OpenRig lets you define a persistent team of Claude Code and Codex agents in YAML, run them under one lead agent, and keep shared context and work organized instead of juggling separate terminal sessions. Aimed at developers running multi-agent coding setups who want coordination, not chaos.
A harness that orchestrates many AI agent harnesses at once — Raven acts as a Host Agent that wires together built-in and third-party agents (Claude, Codex, ChatGPT and others) into DAG-based workflows for complex tasks. It targets recursive self-improving, multi-agent setups rather than a single-model harness, aimed at developers building persistent, self-evolving agent ecosystems.
Google turns raw research notes into submission-ready LaTeX papers — PaperOrchestra, from Google Cloud AI Research, coordinates specialized agents for outlining, literature review, section writing, refinement, and plotting to convert unstructured ideas and experiment logs into submission-ready LaTeX manuscripts. It targets researchers who want to automate the slog from raw notes to a conference-formatted paper, including literature synthesis and generated figures.
A Broadcast Network Built for AI Agents to Talk to Each Other — EigenFlux is an open-source communication and broadcast network designed specifically for multi-agent AI systems. It offers a live feed and dedicated infrastructure for agents to exchange messages, aiming at teams building interconnected AI agent networks rather than single-agent tools.
AWS Labs' orchestrator runs rival coding CLIs as cooperating agents — CAO runs a local cao-server that launches provider CLIs like Claude Code, Codex, Kiro, Copilot CLI, and more inside isolated tmux sessions, letting a supervisor delegate tasks to them in parallel or sequence. Each agent keeps its native authentication and capabilities as a full CLI process, making it useful for teams juggling multiple AI coding assistants.