openhands-ai

What it is and what it does

OpenHands is a full-stack developer control center that orchestrates coding agents—either the open-source OpenHands agent or third-party agents like Claude Code and Codex—across multiple execution backends (local, Docker, VMs, cloud infrastructure). It provides a web UI for starting agent conversations and a REST API backend that manages agent lifecycle, state, and communication. The platform emphasizes flexibility: you can run agents on your laptop, a dedicated machine, or cloud infrastructure, and switch between backends without losing context.

Beyond interactive agent control, OpenHands includes an Automation Server layer for scheduling agents or triggering them via webhooks—enabling workflows like publishing reports to Slack or decomposing GitHub issues into tasks. It integrates with third-party services (Slack, GitHub, Linear, Notion) and supports any LLM. The architecture separates the frontend UI from the backend agent server, allowing distributed deployments where teams can share a central Agent Server for code review while running personal agents locally. With 85 runtime dependencies spanning async frameworks, cloud SDKs, container orchestration, and LLM clients, it is designed as a production orchestration platform rather than a lightweight agent library.

Use it for:

• Run a persistent coding agent on a shared team server for routine tasks while keeping personal agents on your laptop.
• Automate report generation and publish results to Slack on a schedule or via GitHub webhook events.
• Decompose GitHub issues into subtasks automatically and sync them to Linear or Notion.
• Experiment with different LLMs and agent implementations by switching backends from the same UI.
• Deploy agents across multiple cloud providers or on-premises infrastructure from a single control center.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

OpenHands is a self-hosted developer control center that runs coding agents (OpenHands, Claude Code, Codex, Gemini, or any ACP-compatible agent) across local, remote, and cloud backends, with built-in support for automations and integrations.

Yes, if you need a self-hosted, multi-backend agent orchestration platform with automation and third-party integrations. The active maintenance, large community (84043 stars), permissive MIT license, and low install friction support production use. The 85 dependencies and requirement for Python 3.12 or 3.13 are expected for a full orchestration system. No known vulnerabilities. Install if you're building agent-driven workflows; skip if you need a lightweight agent library or are not ready to manage a backend service.

Install

openhands-ai on PyPI

pip install openhands-ai

uv add openhands-ai

poetry add openhands-ai

Installing openhands-ai

Before you install

Low install friction with a pure-Python wheel. Active maintenance (last commit 2026-08-14, 84043 stars). Requires Python 3.12 or 3.13. Runtime dependency footprint is large (85 dependencies including Docker, Kubernetes, cloud SDKs, and async frameworks), reflecting its role as a full orchestration platform rather than a lightweight library.

License in practice

MIT license permits commercial and private use, modification, and distribution with minimal restrictions—standard permissive terms suitable for both self-hosted deployments and integration into other projects.

Quickstart

npm install -g @openhands/agent-canvas
agent-canvas

# Or via Docker:
export PROJECTS_PATH="$HOME/projects"
mkdir -p "$PROJECTS_PATH" "$HOME/.openhands"
docker run -it --rm -p 8000:8000 \
  -v "$HOME/.openhands:/home/openhands/.openhands" \
  -v "${PROJECTS_PATH}:/projects" \
  ghcr.io/openhands/agent-canvas:1.0.0-rc.11

Requires Node.js 22.12.x or later and uv (for local install); Docker Desktop or Engine for sandboxed execution; Python 3.12 or 3.13 for the backend agent server.

Verify before relying

• Whether the 85 runtime dependencies are all required for minimal deployments or if many are optional for specific backends/integrations.
• Performance and resource requirements when running multiple agents or automations concurrently.
• Security hardening specifics beyond the self-hosting guide reference (e.g., network isolation, credential handling).

Package facts

Evidence: openhands_ai-1.11.0-py3-none-any.whl

Tags

More Artificial Intelligence packages

Further reading