Handle engineering requests in Slack
Watch Slack channels for @openhands mentions, open a conversation with the message context, and reply when the agent finishes.
Build it once. Let agents keep it running.
Start with a template or build your own. Trigger workflows from GitHub, Slack, Jira, Linear, schedules, and supported integrations so engineering work keeps moving without someone manually starting every task.
Team communication
Triage every new Linear issue
Classify new Linear issues, suggest labels, find duplicates, and ask clarifying questions before work starts.
Triage incoming Jira incidents
Watch for new bugs or incidents, gather logs and recent changes, suggest severity, and route the issue to the right owner.
Review every pull request
Watch for a configurable label on pull requests, inspect full PR context, and post an AI review comment once per label event.
Draft every incident retrospective
Pull incident notes from Notion and Jira into a shared timeline with owners, decisions, and follow-up tasks.
Remediate security findings
Review security alerts, validate the finding, update affected code, and open a pull request with the remediation.
Fix CI failures automatically
Detect failed workflows, inspect logs, identify the likely cause, and open a pull request with a proposed fix.
Automations
From signal to shipped change.
Turn events from your engineering stack into end-to-end agent workflows. Trigger the work, let agents plan and execute, validate the result, ship the change, and monitor every step.
Signal
Plan
Execute
Validate
Ship
Monitor
Use cases
Put the repetitive parts of the SDLC on autopilot.
Use agents for the repetitive, high-volume work across the software lifecycle, while keeping engineers in control of review, approval, and merge.
01
Review PRs & fix CI failures
Review changes, catch issues, fix failed checks, and open follow-up PRs automatically.
02
Triage issues & open draft PRs
Turn incoming requests into scoped work, gather context, and open a draft PR with the right details attached.
03
Automate QA across the release cycle
Use agents to run repeatable checks, investigate failures, and keep quality work moving from code change through release.
04
Investigate incidents & propose fixes
Pull together alerts, logs, and recent changes, then investigate the issue and propose a patch while humans handle judgment calls.
05
Keep docs & codebase maps current
Update documentation and system context as the code changes so knowledge doesn’t drift out of date.
06
Automate recurring engineering chores
Handle dependency updates, QA sweeps, release notes, and other scheduled work even when laptops are closed.
Integrate with the apps you rely upon
Trigger work from GitHub, Slack, Jira, Linear, and more. Use MCP to connect internal tools, APIs, and data sources so agents can act across the full engineering workflow.
Multi-backend
Your agents. Your models. Your runtime.
Standardize how agent workflows run without standardizing on one agent or model. Use Claude Code, Codex, Gemini CLI, OpenHands, or your own stack, then run those workflows locally, in the cloud, or inside your VPC.
Agent
Runtime
Built for the infrastructure you choose.
OpenHands works with leading AI infrastructure providers to support agent workloads across the models, hardware, and deployment environments your organization already uses.
“The integration of Lemonade Server with OpenHands reflects AMD’s commitment to open-source innovation and developer choice. This collaboration enables local coding agents that prioritize privacy, cost efficiency, and flexible model selection – while taking advantage of acceleration on Ryzen AI PCs. Together, we’re supporting a range of workflows and staying true to open-source values.”
Adrian Macias
Director, Developer Acceleration at AMD
Start on your laptop. Scale to the cloud.
Start locally with the agents and models you choose. Turn what works into shared automations, then scale successful workflows into production.
Individual developer
Start running agents your way.
- Run locally against real code
- Bring Claude Code, Codex, OpenHands, or the agent you prefer
- Run multiple agents in parallel
- Turn useful tasks into your first automation
- Move execution to a VM when your laptop becomes the limit
Team
Turn agent wins into shared automations.
- Share automations instead of rebuilding prompts
- Trigger work from GitHub, Slack, Jira, CI, or schedules
- Keep agents running when laptops are closed
- Run workflows across repositories and teams
- Standardize the patterns that actually work
Enterprise
Run production agents with full control.
- Deploy in your VPC or controlled environment
- Isolate agent execution in secure sandboxes
- Define who and what agents can access
- Audit every action, workflow, and user
- Manage usage, budgets, and organizational policies
Agent Control Plane
Control how agents operate in production.
As agent usage spreads across teams, OpenHands gives platform and security teams centralized control over who can run agents, what they can access, where they execute, what they spend, and what they change.
Developers love OpenHands
Join millions of developers building for the agentic SDLC
Nacho Martínez Rincón
Data Scientist Advocate at Oracle
The OpenHands SDK is one of the most complete SDKs for agent harness and agentic AI development - including enabling vibecoding with opensource models. Great reliability for tool calling through the Tool System - very dev friendly!
Dan Kempe
@danielkempe
OpenHands raised $18.8M to build cloud coding agents that solve 87% of bug tickets same day.
This is what AI coding tools should be. Not another chat interface. Actual autonomous problem solving that ships to production.
Esteban Puerta
@Esteban_Puerta9
At this point, there isn't much of a need to roll your own especially when claude code SDK and openhands sdk are offering nice API's to utilize them.
This is where it gets interesting.
Fisherman
@Rybens92
Your agents are so underrated, but even though there are many benchmarks, which are testing agentic coding and are using your work as reference - OpenHands agent with tested model...
I just don't get why everybody is so silent about your work...
Reposting till people wake up.
Jingyi Zhao
Software Engineer at Walmart Global Tech
I love this project because it is the first of its kind in coding agent space and you have the freedom to configure your own way. It also integrates well with various ecosystems. I believe this project can go long!
Leonardo Gonzalez
@archimagos
I love using autonomous agents like @allhands_ai OpenHands for 80% of the work (esp. backend), and then an interactive IDE like @windsurf or @zeddotdev for debugging, refinement, and UI work.
It saves tons of my time and saves a lot of money too with the right models/plans.
Wil - e/acc
@wilfullyapt
I fvkn love vibe coding with OpenHands. It's not like your editor is doing things for you, it's like an idiot internal that literally knows all the book stuff. OpenHands commit, my tweaks, rinse and repeat. Phenomenal!!
Christian Genco
@cgenco
I haven't tried Devin yet but I freakin' love OpenHands. I prefer having more control over how it's working.
Also that general workflow is the future. Just opening GitHub issues and the AI figures it out and writes tests and then pushes a PR is magical.
OpenHands is an AI agent platform for software development that can execute real engineering work, not just suggest code. It runs autonomous agents that plan, write, and apply changes across your codebase, helping teams complete complex tasks end-to-end. Built on an open source foundation, it's designed for enterprise software development at scale.
Instead of helping you write code faster, OpenHands helps you ship changes end-to-end. Tools such as GitHub Copilot or ChatGPT generate code suggestions. OpenHands runs agents that complete entire engineering tasks, taking actions across entire codebases, running tasks in parallel, and executing changes in real environments.
Yes. OpenHands is specifically designed to handle large, complex, and legacy codebases. With its Large Codebase SDK, it maps dependencies across the system and orchestrates changes in the correct order, allowing multiple agents to safely work in parallel without conflicts.
OpenHands runs in your environment, whether on-prem or private cloud, so your code never leaves your control. Since agents operate inside secure, isolated execution environments with full auditability, it is built for organizations with strict data security, compliance, and governance requirements.
OpenHands is used by platform teams automating engineering workflows at scale, teams building internal developer platforms or AI-powered tooling, and enterprises managing large or legacy codebases and mission-critical software systems.
Open source at the core.
Models change. Agents change. Requirements change. OpenHands gives you an inspectable, extensible foundation you can run locally, self-host, and carry into production without rebuilding around a proprietary platform.
85.3K
GitHub stars
565
Contributors
MIT
Core license
Building the open standard for autonomous software development.
OpenHands is the foundation for secure, transparent, model-agnostic coding agents - empowering every software team to build faster with full control.
