OpenCompany
An open-source OS for self-improving AI employees that run on your own computer and turn tokens into work.
- Source repo
- zeenie-ai/OpenCompany
- Stars
- ★ 1.1k
- Last updated
- 1d ago
- License
- MIT
- Primary language
- Python
- FA score
- 64/100 · Some gaps
At a glance
- How it runs
- Works with
- Universal · cross-platformOpenAI API · Claude API
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- A solo developer or small team hiring a Builder: it writes and runs code, keeps dev servers alive, opens pull requests, deploys, and manages cloud resources via GitHub, Vercel, Cloudflare and Google Cloud.
- Not a fit if
- Teams that need an iOS mobile workspace (only Android emulator on Windows x64 is supported)
- Teams unwilling to store API keys locally or to run self-hosted infrastructure
- Individuals wanting a ready-made hosted chatbot with no connector setup
- Website
- opencompany.sh ↗ set on GitHub
- Maintenance
- Commits in the last 30 dayslast commit 2026-11-20latest release v0.3.2 (2026-11-12)+2.0 ★/day over 7 days
- Source review
- 64/100 · Some gaps
What does this agent do, and when should you use it?
OpenCompany is an MIT-licensed operating system for AI employees that you install on Windows, macOS or Linux, or self-host with the install script, Docker Compose or from source. An employee is not one chatbot but a small team: a lead that understands the job, dispatches work, checks results and reports back, plus specialist agents wired together on a drag-and-drop canvas with no code. Employees connect to Gmail, Google Calendar, Drive, Sheets, Tasks, Contacts, Microsoft 365, WhatsApp, WhatsApp Business, Telegram, Discord, X, Stripe, GitHub, Vercel, Cloudflare, Google Cloud, an Android emulator, a browser, search, scrapers and a knowledge base — 140+ tools in total — and wake on new mail, customer messages or a schedule. Models come from OpenAI, Anthropic, Google, xAI, DeepSeek, Kimi, Mistral, Groq, Cerebras, Sarvam and OpenRouter, or run locally for free through Ollama, LM Studio or any OpenAI-compatible server. Improvement is never model retraining: it lives in editable memory, notes and plain-text skills (78 shipped), plus a live browser workspace and a Windows-x64 Android workspace.
The flow is Hire → Build the team → Start → Review. On Home you describe a job in plain words, OpenCompany drafts the employee's setup (which apps it uses, when it works, what it checks with you first), you adjust and press Hire; alternatively you switch to Dev mode and drop an AI Employee onto the canvas, connect specialist agents and give each the tools for its part of the job (email, browser, code, messaging, payments). Teams run in the background and are event-driven: a new email, a customer message or a scheduled time wakes them, the lead reviews every result from the team and sends it back if it is not right, and you can require approval before anything goes out. Output lands in the employee's Workspace, where you read what it produced and edit what it remembers; skills are short plain-text playbooks that take effect on the next turn and are managed under Settings > Skills. The runtime also ships a live browser workspace — Home and Dev mode can watch an employee browse, let you take control to sign in, then hand it back, all inside a dedicated OpenCompany profile of your installed Chrome, Edge or Chromium — plus a local, controllable Android emulator workspace on Windows x64, and speech and translation so employees can listen, talk and work in other languages.
- A solo developer or small team hiring a Builder: it writes and runs code, keeps dev servers alive, opens pull requests, deploys, and manages cloud resources via GitHub, Vercel, Cloudflare and Google Cloud.
- A community or content operator needing a Grower: it publishes to X, WhatsApp channels, Telegram and Discord on a schedule, while an analyst does social-media data and web research on the market.
- Someone with a heavy inbox running a Chief of Staff on Gmail or Microsoft 365: it reads and answers mail, keeps the calendar, orders files, sheets, tasks and contacts, and delivers a morning summary.
- A customer-facing team using a Front Desk that replies on WhatsApp, WhatsApp Business, Telegram and Discord in the customer's language (voice optional) and escalates what it cannot resolve.
- Research work handled by a Researcher: it browses, searches, scrapes, reads documents and images, and files findings into a knowledge base the whole company can query.
- A payments workflow run by a Treasurer: it operates through Stripe and reacts the moment a payment event fires.
Key setup facts
Pulled from the install and usage notes below. The project README has the current values.
- Install / run
curl -fsSL https://opencompany.sh/install.sh | bash # macOS / Linuxdocker compose up -d --build
- Ports
5678- Config files
~/.opencompany- Runtime
- Python 3.12
- Install notes cover
- Windows · macOS · Linux · Docker
How do you install or deploy this agent?
Desktop app (recommended): download the package for your machine from releases/latest — the Windows x64 .exe, the macOS -arm64.dmg or -x64.dmg, or the Linux .AppImage / .deb. On first launch the app sets itself up (a one-time download of a minute or two). Builds are not code-signed yet, so macOS requires allowing the app under System Settings > Privacy & Security and Windows SmartScreen needs "More info > Run anyway".
Terminal install for servers and headless machines:
curl -fsSL https://opencompany.sh/install.sh | bash # macOS / Linux
iwr -useb https://opencompany.sh/install.ps1 | iex # Windows PowerShell
company startThe script installs bun, Python and uv when missing, then the @zeenie-ai/opencompany package; data lives in ~/.opencompany, shared with the desktop app.
Docker self-hosting:
git clone https://github.com/zeenie-ai/OpenCompany.git
cd OpenCompany
docker compose up -d --buildThat builds the image from source and starts one container with data in the opencompany-selfhost_data volume; open http://localhost:5678 and register the owner account (the port is published on this machine only). Running from source needs bun 1.4+ and Python 3.12:
git clone https://github.com/zeenie-ai/OpenCompany.git
cd OpenCompany
bun run build
bun run devHow do you use this agent?
On first launch the app opens Home; connect an AI provider there or under Settings > Connectors, then describe a job in plain words to hire your first employee, or start from a ready-made bundle in Settings > Plugins. Switch to Dev mode for the workflow editor, where three example employees sit in the workflow sidebar and can be opened to see how they are assembled; drop an AI Employee onto the canvas, connect specialists and configure each one's tools. Watch running teams from their card on Home or the canvas, open the Workspace to see what they made, and change behaviour by editing memory or the plain-text skills; on the Talk page you can ask an employee to take on something new and it can add a tool or learn a skill within the rules you hired it with, taking effect in that conversation immediately and across all its work once you press Apply. Hand control of the browser workspace back and forth when a sign-in is needed. For shared or cloud use, add a login; one command deploys to Google Cloud.
What are this agent's strengths and limitations?
- The team structure is the product, not a prompt: a lead understands the job, hands out work, reviews every result and sends it back if it is not right before it reaches you.
- Model-provider neutral by design — OpenAI, Anthropic, Google, xAI, DeepSeek, Kimi, Mistral, Groq, Cerebras, Sarvam and OpenRouter, plus free local inference via Ollama, LM Studio or any OpenAI-compatible server such as llama.cpp or vLLM.
- Fully inspectable learning: memory, notes, skills and canvas pieces are readable and editable, and the project states it never retrains a model, so there is no black box.
- Improvement happens at runtime: a tool added on the Talk page works in that conversation right away and, with Apply, becomes part of all the employee's work without redeploying.
- MIT-licensed with three documented delivery paths — signed-less desktop installers, the install script, and Docker Compose — and desktop and terminal installs share the ~/.opencompany data directory.
- Desktop builds are not code-signed yet, so macOS and Windows SmartScreen require manual approval — friction for fleet distribution and compliance review.
- Connector and provider presets are deliberately narrow; the README says a new first-class preset needs a reason beyond "my service could be in the dropdown too", so niche SaaS has to arrive via the Apify
customoption, TikHubcall, or a self-saved OpenAI-compatible endpoint. - The mobile workspace is limited to a local Android emulator on Windows x64 and requires accepting the SDK license; iOS is explicitly not supported yet.
- The software is free but real use generally depends on paid model API keys or local inference hardware, and using the 140+ tools means supplying accounts and credentials for each service.
- There is no vendor-hosted product as the main path: self-hosting means Docker or the script and operating it yourself, and running from source adds two runtime prerequisites, bun 1.4+ and Python 3.12.
How does this agent compare with similar options?
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| OpenCompany This agent | 64 · Some gaps | Desktop appFree + model costs | ★ 1.1k | 1d ago | Python | OpenAI API · Claude API |
| Brigade | 53 · Major gaps | Self-hosted serviceFreemium | ★ 16 | today | TypeScript | ChatGPT · Codex · Claude Code · Claude.ai |
| Tale — Shared Workspace for Teams and AI Agents | 65 · Some gaps | — | ★ 47 | 5d ago | TypeScript | Codex · Claude Code · OpenAI API · Claude API |
| BagIdea Office | 46 · Major gaps | Desktop appFree + model costs | ★ 264 | 20d ago | JavaScript | Claude Code · OpenAI API |
How does FollowAgents rate this agent?
Why each dimension lost points
The README says users choose employee access, can monitor work, and can stop or resume employees; it describes asking before sending by default and holding supported sends as drafts for approval. The Android skill tells users to enter passwords themselves and not disclose them in chat. The README also describes local data, encrypted credentials, and the boundary where cloud providers process data. These are useful controls and disclosures, but the supplied source does not show their full implementation or the complete sensitive-data lifecycle. Rollback is limited to stop/resume and applying pending changes, with no general undo or recovery account. Author information appears in package metadata and copyright notices, but maintenance ownership and a trusted update path are not clear.
The CLI bootstrap documents private virtual-environment provisioning, a re-execution guard, and a user-facing failure message; backend startup code logs security-posture warnings. The README links operating modes to architecture documentation. package.json and pyproject.toml agree on version 0.3.1, but this sample cannot verify every documented behavior and contains no execution evidence, so scores remain moderate.
The README serves both nontechnical users and technical operators, with varied employee roles, entry points, cloud and local models, and desktop, server, and Docker deployments. It states that capabilities depend on connected apps, tools, models, and instructions, and clarifies that improvement does not retrain the underlying model. Permission boundaries and trigger conditions are mostly described at a high level. The Android skill specifies a shared local phone and names a Windows x64 limitation, but the sample does not establish environment requirements for every capability.
The README organizes user onboarding, technical operation, architecture, contribution, and licensing, with links to internal guides. Installation directions cover desktop, source, server, and Docker, and disclose unsigned builds and an Android environment limitation. Skill files include tool parameters and examples, though examples and FAQs are not comprehensive. The MIT license file and metadata are present. The supplied materials show a version but no visible changelog, and do not clearly assign maintenance responsibility.
Product documentation describes employee roles, task flow, progress views, draft approval, and workspaces; skill examples show tool inputs and expected responses, making intended outputs understandable for ordinary use. Multi-agent coordination, integrations, and memory may provide practical value, but the materials do not quantify benefits, costs, or resource requirements. Provider charges and local hardware needs are only summarized.
The README links security, installation, and architecture claims to repository documentation; CLI source shows environment provisioning and error handling; skills provide parameter and response examples. This gives some cross-source support. Many product capabilities remain documentation claims, and the supplied source does not demonstrate each implementation. Statements about cloud processing and encrypted credentials are identifiable as product-documentation claims but cannot be independently verified from this sample.
- This is a low-confidence assessment of the supplied static sample; no program was executed, and files not provided are unknown rather than absent.
- The product can access browsers, files, terminals, connected apps, and an Android phone; review permissions and external-send rules for each intended job before deployment.
Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
v0.3.2 OpenCompany v0.3.2
Release notes excerpt
A reliability release. The workflow engine moves to a pinned Temporal server, employees on WhatsApp, Discord and Telegram pick up again after a restart, one large email no longer stops a mail employee, and npm installs can run JavaScript and TypeScript code and read PDFs. The macOS app is signed, so macOS no longer reports it as damaged. Read the upgrade notes first: every employee needs Reset and Start once. Upgrade notes - Reset and Start every employee once. On its first start this release replaces the workflow engine's database (temporal.db, with its -wal and …View on GitHub ↗ -
v0.3.1 OpenCompany v0.3.1
Release notes excerpt
A security and fix release. Dependencies with published advisories are updated, among them brotli, Pillow, pypdf, aiohttp, certifi and Electron, specialists in an employee team keep their settings on every triggered run, and the model catalogue adds Claude Haiku 5.5 and Gemini 4 Argon. Security - brotli 1.2.1 fixes an out-of-bounds write in the Brotli decoder (CVE-2026-111497, "Node2Shell", CVSS 9.2) that a compressed HTTP response could trigger. - Pillow 13.0.0 fixes 39 advisories, four of them critical: heap overflows in the JPEG decoder and the TIFF writer, stale image memory …View on GitHub ↗ -
v0.3.0 OpenCompany v0.3.0
Release notes excerpt
The first release with Home: describe a job, hire an AI employee, talk to it and watch it work. It also brings a new chat, a Browser that drives your installed Chrome or Edge, custom MCP connectors, an Android phone workspace and AWS deployment. - Home: the new start screen. Describe a job, review the setup it drafts and press Hire. Employees can be started, stopped, renamed, given a photo or deleted. Dev mode (Ctrl/Cmd+Shift+D) is the workflow editor. - Talk: chat with each employee on its page. With Ask first on, anything an employee sends waits for your OK, with a 5-second Undo. …View on GitHub ↗
FAQ
Does OpenCompany cost anything beyond model API usage?
Which accounts and keys do I need before the first employee works?
Can an employee send messages or spend money without my knowledge?
What happens across restarts or long idle periods?
Does my data stay on my machine?
Which port does OpenCompany use?
Is OpenCompany open source?
Is OpenCompany safe to use?
Is OpenCompany still maintained?
Who makes OpenCompany?
What is OpenCompany's official website?
Thanks — we use this to improve the page.