What is OpenHands?
OpenHands is an open-source AI agent platform that executes software development tasks end-to-end, planning, writing, and applying code changes across repositories. It supports multiple agents and models, running locally, in the cloud, or inside a private VPC. The core platform is MIT-licensed, with additional cloud and enterprise tiers available.
What are the features of OpenHands?
- Multi-agent orchestrationRuns agents such as Claude Code, Codex, Gemini CLI, or OpenHands using closed models or open weights like Kimi, DeepSeek, and Qwen.
- Event-driven automationsTriggers agent workflows from GitHub, Slack, Jira, Linear, schedules, or MCP-connected internal tools and APIs.
- Large codebase support (Large Codebase SDK)Maps dependencies to orchestrate changes in order, allowing multiple agents to work in parallel without conflicts.
- Isolated execution environmentsRuns agent actions inside secure sandboxes with full auditability for who runs agents, what they access, and what they change.
- Agent Canvas interfaceManages conversations, views automation activity, and configures scheduled workflows like nightly security scans or PR triage digests.
What are the use cases of OpenHands?
- Automated pull request reviewWatches for a configurable label on new pull requests, inspects the context, and posts an AI review comment once per event.
- CI failure remediationDetects failed workflows, inspects logs to identify the cause, and opens a pull request with a proposed fix.
- Incident triageMonitors Jira for new bugs, gathers logs and recent changes, suggests severity, and routes the issue to the appropriate owner.
OpenHands Pricing Information:
- Self-hosted open sourceFree download under the MIT core license; run locally or deploy on your own infrastructure.
- CloudAccount creation required; specific pricing and usage limits are not stated in the source.
- EnterpriseContact sales; includes VPC deployment, isolated sandboxes, access controls, auditing, and budget management.
