HomeToolsCategoriesModel watchFuelGoldInsightsDailyHow we vetContact
Submit a tool

OpenHands

An open-source coding agent that works inside a sandbox.

AI AgentsFree★4.3 editor score

What it does

OpenHands gives an agent a browser, a shell and an editor inside a container, then lets it work a task until the tests pass. Every action is logged, so a failed run can be replayed rather than guessed at.

Self-hosting keeps source code inside your network, which matters when repositories cannot be sent to a vendor. Expect tuning: out of the box it is capable and uneven in equal measure.

Standout features

  • Sandboxed shell and browser
  • Runs locally or self-hosted
  • Full action logs
  • Model-agnostic setup

Best for

AgentOpen SourceSandbox