omnigent-ai/omnigent

Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device.

What it solves

Omnigent provides a unified orchestration layer (a "meta-harness") for various AI agents, allowing users to manage multiple agents from different vendors (like Claude Code, Codex, and Cursor) in a single session without needing to rewrite code for each. It solves the problem of agent fragmentation by enabling cross-agent collaboration, device-agnostic access (terminal, browser, phone), and centralized governance.

How it works

Omnigent acts as a common interface that wraps existing agent runtimes. It supports various model providers (API keys, subscriptions, or gateways like OpenRouter and Ollama) and can launch agents in isolated cloud sandboxes (such as Modal, E2B, or Kubernetes) to ensure security and environment consistency. It uses a local or deployed server to sync sessions across devices and provides a web UI for interaction.

Who it’s for

It is designed for developers and teams who use multiple AI coding agents and want a single place to supervise, collaborate on, and govern their agents' actions, as well as those who want to run agents in secure, disposable cloud environments.

Highlights

  • Multi-Agent Orchestration: Mix and match agents from different vendors in one session to have them review each other's work or split tasks.
  • Cross-Device Sync: Start a session in the terminal and continue it on a mobile phone or browser.
  • Cloud Sandboxing: Run agents in disposable environments via providers like Modal, E2B, and Daytona.
  • Governance Policies: Set rules to pause for approval before risky actions, cap spending, or limit tool access.
  • Real-time Collaboration: Share sessions with teammates to co-drive or fork conversations.

Related

  • Project
    takahirom/arbigent

    Arbigent is an open‑source framework for testing AI‑driven agents on mobile, web and TV apps. It lets you write natural‑language goals, breaks them into dependent scenarios, runs them on real or emulated devices, and validates results with AI‑powered image assertions. Features include cross‑platform support, Maestro YAML integration, reusable scenario libraries, and extensibility via custom hooks and Model Context Protocol servers. Designed for both QA engineers (via a visual UI) and developers (via YAML/CLI), it aims to make AI‑agent testing more maintainable despite current speed and cost constraints.

  • Project
    zhayujie/CowAgent

    CowAgent is an open‑source, self‑hosted AI assistant that plans tasks, runs tools, keeps a personal knowledge base, and can operate as a team of cooperating agents. It supports many LLM providers, multimodal input, dozens of chat channels (Telegram, Slack, WeChat, etc.), a plug‑in skill system, and a three‑tier memory with nightly “Deep Dream” distillation. Install via a one‑line script or Docker, manage everything through a web console, and extend the system through the public Skill Hub or custom tool integrations.

  • Dispatch
    drawgent – AI coding agent on a live Excalidraw canvas

    drawgent lets Claude, Codex or opencode act as a coding assistant that edits an Excalidraw whiteboard in real time, turning on‑canvas notes into live diagram changes.

  • Project
    omnimind-ai/OmniInfer

    OmniInfer is a cross‑platform inference engine that lets you run LLMs and vision‑language models locally via a simple CLI and OpenAI‑compatible HTTP server. It supports many backends (llama.cpp, MLX, MNN, etc.), works on desktop, Android and iOS, and provides hardware‑aware optimisation and a terminal UI for chatting.

  • Project
    agnt-gg/agnt

    AGNT is a local‑first, self‑hostable desktop app that bundles AI agents, visual workflows, long‑running goals, memory, plugins and a local HTTP API into one “agent operating system”. Built with Electron, Vue, Express and SQLite, it supports 15+ LLM providers, 60+ workflow nodes, plugin templates and SkillForge for self‑improving skills. Ideal for developers or teams that need durable, inspectable AI automation on‑premise.