OpenHands (formerly OpenDevin) is a fully autonomous open-source AI coding agent. MIT-licensed core, plans + executes Python actions via CodeAct, uses browser + terminal + files. ~72% on SWE-bench Verified with Sonnet 4.5 via the OpenHands Software Agent SDK (arXiv:2511.03690).
Agenticness = how independently a tool can take action, scored across 9 dimensions. Scored independently by David Kooi, Skylark Creations — see full rubric → Last re-evaluated 31 July 2026.
Is this your tool? Claim this listing to manage your content and analytics.
What it decides by itself, and when it asks you
Mapped from the code in OpenHands/software-agent-sdk.
On its default settings, the software that runs the OpenHands agent never stops to ask for your approval. An app built on it can turn approvals on, but deciding when the work is done and what to remember stays with plain code and AI models.
Does it ask you before it runs a command or edits a file?
Decided by: Plain codeOnly if the app turns on an approval policy; the default policy never asks.
If you send a new message instead of answering, does that count as a yes?
Decided by: Plain codeIt can: when the conversation runs again, it attempts the waiting actions before it looks at your new message, without asking again.
When a helper agent it starts pauses for approval, and the app gave it no approval check of its own, who answers?
Decided by: NobodyNobody: the helper is run again and attempts its waiting actions without asking anyone.
How does it decide it has finished?
Decided by: Plain codeA reply with visible text and no request to use a tool marks the run finished.
Three problems the map found
A new message can count as approval
Fix proposedWhen the approval policy calls for it, the agent pauses before an action until the app approves or rejects it. If you send a new message instead, and the app then runs the conversation again without rejecting, the waiting action is attempted as if you had approved it. A message saying "wait, not that" does not by itself cancel the action.
The goal judge can read "false" as done
Fix proposedWith the goal feature, a judge model is asked for a short structured reply that includes a "complete" field. The code turns that field into yes or no with a Python conversion (bool) that counts any non-empty text as yes, including the word "false". So if the judge writes "false" as text in that field, the goal is marked complete.
A secret inside a tool request is recorded unmasked
ReportedYou can register secrets, such as service access keys, so they are masked in text replies that contain no request to use a tool. When the model puts a registered secret into a request to use a tool, such as a shell command, the details of that request are written to the conversation record without masking, and when the model gives no summary of the request, the automatic summary repeats them. Anyone who can read that record can read the key.
Not checked: This card was taken from the published map of the OpenHands agent SDK at one commit; it did not look at the OpenHands app or its screens, at the hosted services the SDK can call, or at code outside the map's folders, and no OpenHands code was run to make it.
Mapped by agent-ops-patterns (open source) from OpenHands/software-agent-sdk at commit aae9c43. Map dated 2026-10-06.
What's happened with OpenHands lately
- EvaluatedInitial agenticness evaluation: 18/32
News mentions sourced from our news feed; score changes from periodic re-evaluations.
Ask about OpenHands
Get answers based on OpenHands's actual documentation
Try asking:
About
OpenHands is an open-source autonomous coding agent led by Robert Brennan, Xingyao Wang, and the OpenHands community. The agent plans, generates Python actions via the CodeAct pattern, executes them across browser, terminal, and file tools, and...
Proposes and executes multi-step plans with your approval.
Dimension Breakdown
Categories
Ask about OpenHands
Try asking:
- Local (open source, MIT): free. You run it on your own machine and pay your own model provider.
- OpenHands Cloud, Individual: free for one user. Bring your own model key, or use OpenHands' model providers at cost.
- Enterprise: custom pricing, with SSO and the option to run in your own cloud.
When OpenHands changes its pricing, we catch it the next morning
We recrawl every tool in this directory daily. AI Insider is the 3-minute brief on what actually changed — pricing, features, score moves on our 36-point rubric — often before any press coverage. Sent daily. Skipped when nothing happened.