Our take
Activeregular commits and releases- Activity
- 86
- Community
- 93
- Issues
- 83
- Pull requests
- 83
+2,472 stars in the last 30 days
- Merges more pull requests than 94% of the projects we track
- Releases more often than 89% of the projects we track
- More contributors than 78% of the projects we track
OpenHands Agent Canvas is a actively maintained, MIT-licensed control center for teams that want to self-host and orchestrate coding agents, though its beta status and direct-machine access options call for careful deployment choices.
Good fit if
- You want one interface for agents running across local, remote, and cloud backends.
- You need scheduled or event-triggered engineering automations connected to services such as Slack or GitHub.
- Your team can operate a Node.js 24-or-later environment and wants the flexibility of an MIT-licensed project with active development and broad contributor participation.
Look elsewhere if
- You need a standalone coding agent rather than a control center that connects to agent servers.
- You cannot accommodate Node.js 24 or later for the npm or source launchers.
- Your deployment cannot safely grant an agent access to project files or provide a suitable sandbox and network configuration.
All health signals
| Last commit | 2026-10-20 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 66 (latest v1.29.0, 2026-10-20) |
| Contributors | 100+ (top contributor: 14% of commits) |
| Issues closed, last 90 days | 37+ (typically closed in 1 day) |
| Pull requests merged, last 90 days | 91 (typically merged in 0 days) |
| Project age | 2 years |
Checked on 2026-10-21 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
93,483Forks
12,514Change since Nov 14, 2026: +46 stars·+11 forks
Overview
OpenHands Agent Canvas is a self-hosted developer control center built on the OpenHands Agent Server, a REST API from OpenHands for running agents. It gives developers one place to start agent conversations and automate routine engineering tasks, with agents running on local, remote, or cloud backends.
The project is aimed at developers and teams who want to manage coding agents across different environments. It runs the OpenHands agent out of the box and supports third-party agents such as Claude Code, Codex, Gemini CLI, Pi, and OpenCode, as well as other Agent-Client Protocol-compatible agents.
Key Features
- Connect to multiple Agent Servers and switch between local, remote, and cloud backends.
- Run the OpenHands agent or connect supported third-party and ACP-compatible agents.
- Create scheduled or event-triggered automations that integrate with services such as Slack, GitHub, and Linear.
- Run agents directly on a machine, in Docker, on virtual machines, or on company infrastructure.
- Use Docker sandboxes to isolate conversation execution, including separate containers for concurrent conversations.
- Configure the system to use different LLMs through its LLM profiles.
- Launch the frontend and local services together, or run frontend and backend components separately.
Use Cases
- A developer can run a coding agent against local projects and switch to a remote Agent Server when work needs to continue off their laptop.
- A team can host agents on shared infrastructure for recurring code review or dependency-update work.
- An engineering team can trigger task decomposition or other agent workflows from GitHub events or scheduled automations.
- A developer can run concurrent agent conversations in Docker containers when isolated execution is useful.
What you need
Detected in the repository
- Node.js >=24 (from package.json)
- A Helm chart for Kubernetes
- A test suite and automated checks on GitHub Actions
License in plain words
MITpermissive
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the copyright and license notice
- Share your changes: no
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
With Node.js 24 or later and uv installed, start the local stack with:
npm install -g @openhands/agent-canvas
agent-canvas
Open http://localhost:8000. See the README and self-hosting guide for Docker, remote backends, and deployment details.
Alternatives
- tessera: Tessera coordinates local Claude Code, Codex, and OpenCode sessions with Git worktrees, while OpenHands also runs agents on remote and cloud backends.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| OpenHands | TypeScript | MIT | 91.4k | Active |
| tessera | TypeScript | AGPL-3.0 | 314 | Active |
Considerations
- The project is identified as beta, so teams should evaluate its suitability for their deployment and operational needs.
- The unsandboxed local and source launch options run the agent server directly on the host, giving the agent access to that machine's filesystem. Docker sandboxing is available, but requires Docker and does not move the entire Canvas and automation service into a sandbox.
- Network exposure needs deliberate configuration. The README warns that public or LAN-facing installs require appropriate API-key handling and recommends its self-hosting guidance for internet-facing deployments.
- The local and source launchers require Node.js 24 or later and
uv. The repository also depends on the separately maintained OpenHands Agent Server and TypeScript client, so changes to those components belong in their respective repositories.
Found this useful?
Share it with someone who would like OpenHands.