<div align="center">
<img src="docs/assets/readme-banner.webp" alt="AgenticOS: open-source AI teammates for your company. Agents, drawn as blobs, work with the Toolbox's capabilities: knowledge search, a web browser, subagents, a sandbox, charts, apps, skills and guardrails." width="100%">
<p>
<b>Open-source, self-hosted AI agent platform for your company's AI teammates.</b><br>
Build, run and govern agents on your own servers and models: shared chat, RAG, skills, a real browser, MCP, budgets, approvals and an audit trail.<br>
<sub>Apache-2.0 · built on Pydantic AI</sub>
</p>
<p>
🏆 <b>Built by Vstorm — a recipient of the 2026 Open Agentic AI Innovation Award.</b><br>Recognized by the Agentic AI Foundation for adopting AGENTS.md and integrating MCP in its open-source framework.<br><a href="https://aaif.io/blog/winners-of-the-first-aaif-annual-technical-excellence-awards">Read the AAIF announcement</a> · <a href="https://vstorm.co/agentic-ai/open-agentic-ai-standards-aaif-award/">The work behind the award</a>
</p>
<p>
<a href="#-see-it-in-action">See it</a> ·
<a href="#-highlights">Highlights</a> ·
<a href="#-what-is-agenticos">What is it?</a> ·
<a href="#-quick-start">Quick start</a> ·
<a href="#run-on-your-server-open-it-anywhere">Server, desktop & mobile</a> ·
<a href="#-connect-the-apps-your-team-already-uses">Integrations</a> ·
<a href="#-build-share-and-operate">Product tour</a> ·
<a href="#-find-your-path">Find your path</a> ·
<a href="#-what-ships-today">What ships</a> ·
<a href="#-frequently-asked-questions">FAQ</a> ·
<a href="https://vstorm-co.github.io/vstormos-landing/">Website</a> ·
<a href="https://vstorm-co.github.io/agenticos/presentation/">Introduction deck</a> ·
<a href="https://vstorm-co.github.io/agenticos/">Documentation</a>
</p>
<p>
<a href="https://github.com/vstorm-co/agenticos/actions/workflows/ci.yml"><img src="https://github.com/vstorm-co/agenticos/actions/workflows/ci.yml/badge.svg?branch=main" alt="CI"></a>
<a href="https://github.com/vstorm-co/agenticos/releases"><img src="https://img.shields.io/github/v/release/vstorm-co/agenticos?label=release&color=blue" alt="Release"></a>
<a href="LICENSE"><img src="https://img.shields.io/badge/licence-Apache--2.0-blue" alt="Apache-2.0"></a>
<a href="https://ai.pydantic.dev"><img src="https://img.shields.io/badge/Powered%20by-Pydantic%20AI-E92063?logo=pydantic&logoColor=white" alt="Built with Pydantic AI"></a>
</p>
<p>
<b>English</b> ·
<a href="README.pl.md">Polski</a> ·
<a href="README.de.md">Deutsch</a> ·
<a href="README.es.md">Español</a>
</p>
</div>
<a href="docs/assets/readme/highlight-hero.webp"><img src="docs/assets/readme/highlight-hero.webp" alt="Prepared Revenue Pulse sales review in the full AgenticOS chat: revenue vs target and team contribution." width="100%"></a>
*Prepared conversation with fictional sales data in the real Visuals renderer. No live CRM or new model-generation run is represented.*
AgenticOS gives every team AI teammates: agents that work with files, run code, browse the web and use your company's tools and knowledge. Your colleagues and your agents work in the same live chat, and any agent can hand code to Claude Code or Codex, on the ChatGPT or Claude subscription you already pay for. It is open source (Apache-2.0) and self-hosted, a layer you run on your own infrastructure. Build and publish agents in the browser, share them with colleagues, and manage their access, cost and results in one place.
**New here?** Click through the [17-slide introduction](https://vstorm-co.github.io/agenticos/presentation/): the problem, the idea, the product on real screens, its controls and limits, and how to start. For every screen in detail, open the [47-slide product tour](https://vstorm-co.github.io/agenticos/presentation/tour/). Arrow keys step through either deck; `O` lists every slide.
## New in the latest releases
- **Your local Claude Code or Codex, inside desktop chat.** Choose a coding agent per conversation, work in an approved local folder, resume its session and apply corrections while it works. Sign in through the coding agent's own terminal flow.
- **Your skills and project instructions.** Opt a desktop conversation into local skills and `AGENTS.md` / `CLAUDE.md`, read from your computer. Each new chat starts with this sharing off.
- **Discord joins Slack, Teams, Telegram and Mattermost.** Answer mentions, direct messages and `/ask`; review approvals and questions in the channel. [Set up a channel](https://vstorm-co.github.io/agenticos/channels/).
- **More useful Visuals and Apps.** Draw country maps, pins and routes with the bundled `AOWorldMap`, or use the presentation skills to build a deck as an App. A command watcher can follow test or log output while an agent works.
[Release notes](https://github.com/vstorm-co/agenticos/releases) · [Desktop guide](https://vstorm-co.github.io/agenticos/desktop/)
## ⚡ Highlights
### Your team and your agents, in one live room
A project gives everyone the same files, instructions and a computer for the agents. Invite colleagues and work the way a team works: see who is here, watch a teammate's agent answer as it is written, and bring in any agent, tool or document with an `@`.
<a href="docs/assets/readme/highlight-projects.webp"><img src="docs/assets/readme/highlight-projects.webp" alt="A shared project chat with a launch brief and channel budget, shown with the complete console navigation." width="100%"></a>
### Reply to anyone, and the agent knows who said what
Reply to a person or an agent's answer, and the agent applies the change and says whose it was. Every message carries a name, so a decision has an owner.
<a href="docs/assets/readme/highlight-reply.webp"><img src="docs/assets/readme/highlight-reply.webp" alt="Tom Brooks replies to the Chief of Staff's launch plan with a budget change; the agent answers Tom, applies his change and Peter's confirmation, and the Finance Analyst rechecks the numbers" width="100%"></a>
### Claude Code and Codex, on your team
Any agent can hand software work to Claude Code or Codex. They build in the agent's own computer, on your ChatGPT or Claude subscription, and report what they changed, what they checked and what they need from you.
<a href="docs/assets/readme/highlight-coding.webp"><img src="docs/assets/readme/highlight-coding.webp" alt="Prepared interface example: Claude Code implements a supplier portal and Codex reviews it in read-only mode. Fictional handoffs; no coding CLI execution is represented." width="100%"></a>
*Prepared interface example: Claude Code implements a supplier portal and Codex reviews it in read-only mode. Fictional handoffs; no coding CLI execution is represented.*
### Ask how it works, get a drawing you can use
An agent draws while it writes: a diagram from your documents, a calculator with sliders, a chart or a 3D scene, live under its message. One click saves it as an app your team opens by link.
<a href="docs/assets/readme/highlight-visuals.webp"><img src="docs/assets/readme/highlight-visuals.webp" alt="A prepared manufacturing Visual inside chat: a supply network, inbound delivery exceptions and assembly windows. Interactive fictional data; no ERP, machine connection or live model call is represented." width="100%"></a>
*A prepared manufacturing Visual inside chat: a supply network, inbound delivery exceptions and assembly windows. Interactive fictional data; no ERP, machine connection or live model call is represented.*
### Write to an agent and leave
Ask for research that takes a while and close the tab. The agent keeps working, the answer lands in the conversation, and the bell tells you it is done.
<a href="docs/assets/readme/highlight-research.webp"><img src="docs/assets/readme/highlight-research.webp" alt="A Sales Researcher's table of what five AP automation tools charge a 200-person company, with plan, price, what is included and a source link for every row" width="100%"></a>
### Your day, on one screen
Open the console to what your agents did, the calls waiting on you and the conversations to pick up again - with a tour of what they can do that is easy to miss.
<a href="docs/assets/readme/highlight-dashboard.webp"><img src="docs/assets/readme/highlight-dashboard.webp" alt="The personal Me dashboard with an actionable briefing, team knowledge follow-ups and a carousel of AgenticOS capabilities." width="100%"></a>
### Agents that work as a team
A lead agent hands parts of a task to colleagues and writes up what they found - you see who it asked, and what each said.
<a href="docs/assets/readme/highlight-subagents.webp"><img src="docs/assets/readme/highlight-subagents.webp" alt="A Chief of Staff agent's pricing review brief, written from what Sales Researcher and Finance Analyst answered: Ramp and Bill.com pricing in dollars, the margin figure Finance could not give without a P&L, and three bullets for the meeting" width="100%"></a>
### A real browser, right in the chat
Watch it click through sites, take over when a sign-in needs you, point at a spot to comment.
<a href="docs/assets/readme/highlight-browser.webp"><img src="docs/assets/readme/highlight-browser.webp" alt="The Sales Researcher’s recorded browser steps and sourced answer about Ramp pricing." width="100%"></a>
### Build, equip and test in one window
Enable tools and try a real task in the draft test panel before publishing the agent for colleagues.
<a href="docs/assets/readme/highlight-builder-test.webp"><img src="docs/assets/readme/highlight-builder-test.webp" alt="Finance Analyst calculating a sample invoice in the builder’s draft test panel: subtotal €6,180, VAT €1,421.40 and total €7,601.40." width="100%"></a>
### Apps, not just answers
An agent publishes a dashboard or a report as a page with a stable link and versions.
<a href="docs/assets/readme/highlight-app.webp"><img src="docs/assets/readme/highlight-app.webp" alt="Northline manufacturing App: supplier routes, delivery exceptions and an assembly plan." width="100%"></a>
*Northline: a prepared supply and production planning App with fictional shipments, local delay simulation and assembly windows.*
<details>
<summary>More prepared Apps: motor assembly and sales intelligence</summary>
Working local demonstrations with fictional data. Motor Atlas lets you inspect a conceptual assembly and its component records; Revenue Pulse recalculates sales metrics by region and period. No equipment, ERP or CRM is connected.
<a href="docs/assets/readme/highlight-revenue.webp"><img src="docs/assets/readme/highlight-revenue.webp" alt="Revenue Pulse App with sales targets, pipeline and team contribution." width="100%"></a>
</details>
### A website widget, on your own site
The actual AgenticOS widget runs a published, read-only agent on the fictional Trail Supply website. In the recorded local test it answered using the seeded catalog in its instructions. Live inventory, purchases and payments are not connected.
<a href="docs/assets/readme/highlight-widget.webp"><img src="docs/assets/readme/highlight-widget.webp" alt="Fictional Trail Supply website with the actual AgenticOS widget answering a daypack question." width="100%"></a>
[Build and publish a website widget](https://vstorm-co.github.io/agenticos/howto/support-widget/)
### 4,700+ MCP servers
Connect Notion, GitHub, Jira or your own server, and choose which tools each agent may call.
<a href="docs/assets/readme/highlight-mcp-servers.webp"><img src="docs/assets/readme/highlight-mcp-servers.webp" alt="The MCP server gallery with connected servers and their sign-in method" width="100%"></a>
### Work in AgenticOS from Claude Code, Codex or Cursor
AgenticOS is an MCP server. Add it with one command and sign in through the browser: your coding assistant can list and run agents, read runs and what they cost, and search and fill knowledge bases, within what you allowed.
<a href="docs/assets/readme/highlight-mcp-server.webp"><img src="docs/assets/readme/highlight-mcp-server.webp" alt="The MCP clients settings: the AgenticOS server URL, sign-in through the browser or with an API key, and the one-line command that adds it to Claude Code, with tabs for Codex, Claude Desktop, Cursor and VS Code" width="100%"></a>
## 📸 See it in action
Browse the [current product gallery](https://vstorm-co.github.io/agenticos/screens/). Motor Atlas below is a prepared, interactive assembly explorer with fictional component records and telemetry; it is not a CAD drawing.
<a href="docs/assets/readme/highlight-motor.webp"><img src="docs/assets/readme/highlight-motor.webp" alt="Motor Atlas App with an interactive exploded motor assembly and component register." width="100%"></a>
## 🧩 Build useful work around your agents
Use [Workflows](https://vstorm-co.github.io/agenticos/workflows/) to connect typed steps, agent calls and MCP tools on a visual canvas. Store structured results in [Virtual Tables](https://vstorm-co.github.io/agenticos/virtual-tables/), start work with [Routines](https://vstorm-co.github.io/agenticos/triggers/), and review runs and approvals. [Voice calls](https://vstorm-co.github.io/agenticos/voice/) and the [desktop app](https://vstorm-co.github.io/agenticos/desktop/) provide other ways to work with the same published agents.
**Visual workflows, from lead intake to a human-reviewed handoff.**
<a href="docs/assets/readme/highlight-workflows.webp"><img src="docs/assets/readme/highlight-workflows.webp" alt="Illustrative 30-step draft with Google Workspace, Microsoft 365, HubSpot, Salesforce, Notion, Slack, agents and approvals. Connections are demo placeholders; this is not an executed integration." width="100%"></a>
Illustrative 30-step draft with Google Workspace, Microsoft 365, HubSpot, Salesforce, Notion, Slack, agents and approvals. Connections are demo placeholders; this is not an executed integration.
**Structured work your agents and your team can read together.**
<a href="docs/assets/readme/highlight-tables.webp"><img src="docs/assets/readme/highlight-tables.webp" alt="The complete AgenticOS window showing a fictional sales pipeline as a Kanban board." width="100%"></a>
Virtual Tables with fictional companies, typed fields, stages and next actions. Switch between a grid and a kanban board.
Start with a recognizable task: an [LLM wiki in a Project](https://vstorm-co.github.io/agenticos/projects/), a [website agent widget](https://vstorm-co.github.io/agenticos/channels/), or a team working from shared project files and instructions. A widget supplies the conversation surface; recommendations and cart actions still need your catalog, tool integration and permission checks.
## 💡 What is AgenticOS?
**AgenticOS is an open-source (Apache-2.0), self-hosted layer for building, sharing and governing AI agents across a company.** Teams configure an agent in the browser by writing its instructions, choosing a model and switching on tools. They connect it to company documents and apps, and publish it to web chat, Slack, Teams, Discord, Mattermost, Telegram, a website widget or an API. Administrators control who can use each agent, what it may spend and which actions need a person's approval. Every run is recorded.
Most agent frameworks give you a library, so every change to an agent's behaviour is a pull request, a review and a release. That is the wrong shape for the small agents a company actually wants, because the person who knows what the agent should say is not the person with commit access. **Code defines, configuration composes:** engineers extend what there is to assemble, and configuration can only ever reach what code registered.
It runs on your own infrastructure with Docker Compose and works with 28 model providers, including your ChatGPT subscription and local models through Ollama and vLLM. Agents run on [Pydantic AI](https://ai.pydantic.dev) and [pydantic-ai-harness](https://pydantic.dev/docs/ai/harness/); the layer around them uses FastAPI, PostgreSQL with pgvector and Next.js. It is maintained by [Vstorm](https://vstorm.co).
**Who it is for:**
- Companies that want **an internal AI agent layer** they own, instead of per-seat assistants in a vendor's cloud.
- Teams with **repeated work over documents and tools**, such as reports, support answers, contract checks and data analysis.
- IT and security teams that need **data sovereignty, company sign-in, budgets, approvals and an audit trail** for AI agents.
- Engineers who want **typed Python extension points** and a console their non-technical colleagues can use.
<a href="docs/assets/readme/company-architecture-diagram.webp"><img src="docs/assets/readme/company-architecture-diagram.webp" alt="AgenticOS inside your company: departments and systems on the left; AgenticOS with example agents and the controls every request passes in the middle; your data, sandboxes, vault and optional local models inside; hosted models, SaaS tools and document sources outside, only if you choose." width="100%"></a>
<p align="center"><sub><b>How it fits into your company.</b> Illustrated people; the agents are examples.</sub></p>
**How AgenticOS fits into a company:** departments such as finance, operations or the board use shared agents in web chat, Slack or a private workspace. Your systems call agents through the API, and events or schedules start them automatically. Every request passes the same controls: roles, budgets, approvals, guardrails and a recorded run. Your data, vectors, code sandboxes, the credential vault and optional local models stay on your infrastructure. Hosted models, SaaS tools and document sources outside are used only when you configure them.
28 model providers · 4,700+ MCP gallery entries · shared chat and Projects · visual workflows and virtual tables · web, desktop, API and messaging channels. Configure availability and tools for each published agent.
## ✨ What you can do
| For your team | What AgenticOS provides |
|---|---|
| [Work with files and code](#-work-with-files-and-code) | Analyze CSVs, produce charts and documents, work on repositories |
| [Build reusable agents](#-build-agents-your-team-can-reuse) | Choose models and tools, publish versions, share agents with colleagues |
| [Connect company knowledge](#-teach-agents-how-your-team-works) | Reuse skills, context and searchable documents across agents |
| [Share the results](#-publish-results-as-interactive-pages) | Publish interactive pages with stable links and version history |
| [Run and monitor](#-track-runs-costs-and-approvals) | Customize dashboards, inspect runs, schedule tasks and set budgets |
| [Organize company access](#-organize-teams-with-roles-and-groups) | Combine roles, department groups and company sign-in |
## 🚀 Quick start
### Choose how you want to start
**One company workspace, many ways to connect.** The server runs AgenticOS; the desktop app connects to it. Joining an existing workspace needs no Docker or developer tools.
| Your situation | Next step |
|---|---|
| Your company already has AgenticOS | Ask for the workspace URL and sign in through your browser, or [install the desktop app](#install-the-desktop-app). |
| You want a workspace for your company | Have your IT team [deploy a server](#deploy-a-company-server), or [ask Vstorm to help](https://vstorm.co/contact-us/). |
| You want to try it on your computer without terminal commands | [Install the desktop app](#install-the-desktop-app), choose **On this computer** and follow setup. Review the [requirements](https://vstorm-co.github.io/agenticos/desktop/#running-it) first; AI requires a configured model, while recorded examples need none. |
| You are a developer exploring locally | [Run the Docker quickstart](#try-it-locally). |
### Try it locally
<details id="local-quickstart">
<summary>For developers: try AgenticOS locally with Docker</summary>
Start with Docker Compose and access to a model provider. On macOS or Linux, run:
```bash
curl -fsSL https://raw.githubusercontent.com/vstorm-co/agenticos/main/scripts/quickstart.sh | bash
```
On Windows, run the same command inside WSL2 with Docker Desktop's WSL2 integration switched on. The installer asks for a model provider and key, your login and an organization name, pulls the published images and starts a deployment with a working agent in it. A host with 4 vCPU and 8 GB of RAM runs it. Then open **http://localhost:3000** and sign in with the login you chose during installation.
**Your first agent:** follow the [document-assistant walkthrough](https://vstorm-co.github.io/agenticos/howto/first-document-agent/) to upload a handbook, ask questions and check answers against cited sources. Then pick a next task from [30 tutorials](https://vstorm-co.github.io/agenticos/use-cases/), each with a sample input and a check you can run.
<details>
<summary>Inspect the installer or deploy another way</summary>
Read the [installer](scripts/quickstart.sh) before running it. To check prerequisites without installing:
```bash
curl -fsSL https://raw.githubusercontent.com/vstorm-co/agenticos/main/scripts/quickstart.sh | bash -s -- --check
```
For manual Docker Compose setup, pinned versions and troubleshooting, follow the [installation guide](https://vstorm-co.github.io/agenticos/install/). For development from source, see [Contributing](https://vstorm-co.github.io/agenticos/help/).
</details>
</details>
## Run on your server. Open it anywhere.
The server holds the workspace. The browser, desktop app and phone connect to that deployment with its accounts, agents and access controls.
<a id="server-setup"></a>
### Deploy a company server
Your IT team sets up the workspace once, then shares an HTTPS address and creates your account. You can use that address in a browser or in the desktop app. For an assisted rollout, [talk to Vstorm](https://vstorm.co/contact-us/).
<details>
<summary>Server setup for IT and developers</summary>
Use a Linux host with **4 vCPU and 8 GB RAM**, Docker Engine 24+, Compose 2.24+ and `make`. Point two DNS names at it — for example `app.example.com` and `api.example.com` — and allow HTTPS through a reverse proxy. The supported deployment is Docker Compose on one host.
```bash
# On the server. Start from a released version so code and images stay aligned.
git clone --branch v1.38.2 --depth 1 https://github.com/vstorm-co/agenticos.git
cd agenticos
bash scripts/server-init.sh
```
The setup asks for the site/API hostnames and certificate email, then generates `backend/.env`. Set `AGENTICOS_VERSION=1.38.2` there to pin this release. For a new host using the bundled Traefik proxy:
```bash
docker network create traefik_webgateway
docker compose --env-file backend/.env -f docker-compose-traefik.yml up -d
make prod PROXY=traefik
make prod-frontend PROXY=traefik
```
Then [bootstrap the first organization, owner and agent](https://vstorm-co.github.io/agenticos/deploy/#start-it-and-create-the-first-account), open your site's HTTPS address and add your organization's model credentials. The guide also covers an existing proxy, Nginx, updates and backups. Keep a backup of the database **and the vault key in `backend/.env`**.
```bash
docker compose --env-file backend/.env -f docker-compose-prod.yml \
exec -T app agenticos cmd bootstrap \
--email you@example.com --password '<your-password>' \
--org 'Your Company' --provider anthropic --api-key '<your-api-key>'
```
[Complete server deployment guide](https://vstorm-co.github.io/agenticos/deploy/) · [Configuration checklist](https://vstorm-co.github.io/agenticos/configuration/#production-checklist)
</details>
<a id="desktop-install"></a>
### Install the desktop app
Download an installer from [release v1.38.2](https://github.com/vstorm-co/agenticos/releases/tag/v1.38.2):
| System | Download v1.38.2 | What to do |
|---|---|---|
| <img src="docs/assets/readme/platforms/apple.svg" width="18" height="18" alt="" /> **macOS** | [Apple silicon · DMG](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS_1.38.2_aarch64.dmg) · [Intel · DMG](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS_1.38.2_x64.dmg) | Open the DMG and drag AgenticOS into Applications. |
| <img src="docs/assets/readme/platforms/windows.svg" width="18" height="18" alt="" /> **Windows x64** | [EXE](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS_1.38.2_x64-setup.exe) · [MSI](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS_1.38.2_x64_en-US.msi) | Open the EXE and follow the installer. MSI is for managed IT deployment. |
| <img src="docs/assets/readme/platforms/linux.svg" width="18" height="18" alt="" /> **Linux x64** | [DEB](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS_1.38.2_amd64.deb) · [RPM](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS-1.38.2-1.x86_64.rpm) · [AppImage](https://github.com/vstorm-co/agenticos/releases/download/v1.38.2/AgenticOS_1.38.2_amd64.AppImage) | Open DEB/RPM in your software installer. For AppImage, allow execution in file properties. |
On a Mac, choose **Apple menu → About This Mac**: “Chip” identifies Apple silicon; “Processor: Intel” identifies an Intel Mac.
**Prefer Homebrew on macOS?**
The project’s [Homebrew tap](https://github.com/vstorm-co/homebrew-agenticos) chooses the correct Mac installer. Its cask clears the download quarantine flag for AgenticOS; it does not provide Apple notarization:
```bash
brew install --cask vstorm-co/agenticos/agenticos
```
**After installing:** open AgenticOS, enter the HTTPS workspace address your team gave you and sign in. Connecting to an existing server needs no local Docker. To create a local workspace instead, choose the local-instance option in the wizard; that needs Docker.
On macOS, builds are not Apple-notarized; a browser-downloaded DMG may need **System Settings → Privacy & Security → Open Anyway**. See the [installation guide](https://vstorm-co.github.io/agenticos/desktop/#installing-the-app).
Beyond the console, desktop adds Quick Ask, local files and computer access with your consent, and local Claude Code / Codex. Under **Settings → Desktop app**, check the installed coding agents and sign in through their terminal. Choose one in **Chat controls**, grant a working folder full access and confirm the task in the native dialog. **My skills and instructions** is a separate, optional choice for each chat.
[Desktop setup and local agents](https://vstorm-co.github.io/agenticos/desktop/)
### Use AgenticOS on your phone
**Available now without a build:** open your deployment's HTTPS address in a mobile browser and sign in. Use the install prompt on a supported browser; on iPhone, choose **Share → Add to Home Screen**. The console adapts to a phone, including chat, files and approvals.
**Native iOS and Android:** the source is included in `mobile/`, with biometric locking, sharing, widgets and native notification integrations. These are currently **development builds**, with no public App Store, Google Play, TestFlight or supported release download. To try the native app, install Rust plus Xcode on macOS (iOS), or the Android SDK/NDK and JDK (Android), then:
```bash
cd mobile
bun install --frozen-lockfile
bun run ios # iOS development on macOS
# or: bun run android
```
Install with your own platform signing/provisioning where required. Enter a server address reachable from the phone, or pair from **Settings → Mobile app** using the one-time QR code and confirm the account. `localhost` on a phone refers to the phone itself. Native push needs the deployment's platform-specific configuration; the [mobile guide](https://vstorm-co.github.io/agenticos/mobile/) covers build prerequisites, pairing and notifications.
## 🔌 Connect the apps your team already uses
AgenticOS connects agents to the models, chat tools, business apps and document stores a company already uses, so an agent can read a Notion brief, search SharePoint documents or answer in Slack under the same access rules and budget.
<a href="docs/assets/readme/integrations-hub.webp"><img src="docs/assets/readme/integrations-hub.webp" alt="AgenticOS as a hub: models it thinks with on top; where people reach it and what starts it on the left; tools it can use through MCP on the right; documents it reads at the bottom." width="100%"></a>
| Connect | How | Read |
|---|---|---|
| **Chat tools** | Publish an agent to Slack, Mattermost or Telegram, a website widget, a hosted page, the API or a WebSocket | [Channels](https://vstorm-co.github.io/agenticos/channels/) |
| **Business tools** | **4,700+ MCP servers** in the gallery (Notion, GitHub, Jira, HubSpot, Stripe…) and your own servers; choose which tools each agent may call | [MCP](https://vstorm-co.github.io/agenticos/mcp/) |
| **Documents** | Sync Google Drive, S3/MinIO, Git repositories, websites, SharePoint and OneDrive into knowledge bases | [Sync sources](https://vstorm-co.github.io/agenticos/howto/configure-sync-sources/) |
| **Events** | Start agents on a schedule, a new GitHub issue, a Gmail message or a signed webhook | [Routines](https://vstorm-co.github.io/agenticos/triggers/) |
| **Models** | 28 providers, your ChatGPT subscription, your cloud contract (Azure, Bedrock, Vertex) or local models (Ollama, vLLM) | [Models](https://vstorm-co.github.io/agenticos/models/) |
<sub>Registry listings are publisher-provided metadata; each connection needs its own setup and access review. Outlook email and calendar connect through a third-party MCP service. Logos identify connection options and do not imply a partnership.</sub>
## 🧩 Build, share and operate
### 🤖 Build agents your team can reuse
Choose an agent's model, instructions and tools in the browser. Publish a version for colleagues to use; inspect earlier versions and roll back when needed. Keep specialized agents for research, reporting, coding or operations in one catalog.
<img src="docs/assets/readme/builder-annotated.webp" alt="The agent builder with four numbered areas: name and status, tabs, instructions and model." width="100%">
The agent builder has four areas: **(1)** the name and publishing status, where a draft stays private until you publish a version; **(2)** tabs for the toolbox, MCP servers, limits, availability and version history; **(3)** the instructions, written in plain language like a brief for a new colleague; and **(4)** the model, chosen per agent from your configured providers.
Colleagues can use a published agent in **web chat, Slack, Mattermost or Telegram** when those channels are configured, on a **website widget** or a **hosted page**, or through the **API** and **WebSocket**. [Build an agent](https://vstorm-co.github.io/agenticos/first-agent/) · [Connect a channel](https://vstorm-co.github.io/agenticos/channels/)
### 📂 Work with files and code
Ask an agent to analyze a spreadsheet, produce a chart, prepare a document or work on a repository. With a container-backed sandbox configured and command execution enabled, it can **read and edit files, run shell commands, and execute Python or JavaScript**. The bundled workbench includes data, charting and document tools, including LibreOffice.
If you use [Claude Code](https://code.claude.com/docs/en/overview) or [Codex](https://developers.openai.com/codex/cli/), the file-and-command workflow will feel familiar. AgenticOS brings that kind of work into a shared, self-hosted layer with company knowledge, reusable agents and organization access controls. What an agent can accomplish depends on its model, enabled tools and instructions. [Sandbox configuration](https://vstorm-co.github.io/agenticos/sandbox/)
### 🧠 Teach agents how your team works
- **Skills** hold reusable procedures: how to review code, write a report or research a market. Maintain them once and reuse them across agents.
- **Context** holds standing knowledge such as a glossary, policy or brand voice. Include it in the prompt or let the agent read it on demand.
- **Knowledge bases (RAG)** make uploaded documents searchable. Choose parsing options, inspect processing status and chunks, or sync them from a source listed below.
<img src="docs/assets/readme/rag-pipeline.webp" alt="From a file to a cited answer: sources, read, split, embed, answer." width="100%">
**How retrieval-augmented generation (RAG) works in AgenticOS:** documents are uploaded or synced from Google Drive, S3/MinIO, Git, websites, SharePoint or OneDrive. They are read by a parser: PyMuPDF and LiteParse run locally, with OCR for scans, while LlamaParse is a cloud service. Documents are then split into chunks, embedded with an OpenAI, OpenRouter or local Ollama model, and stored in PostgreSQL with pgvector. At question time the agent searches the vectors with filters and answers with citations.
[Skills](https://vstorm-co.github.io/agenticos/skills/) · [Context](https://vstorm-co.github.io/agenticos/context/) · [Document processing](https://vstorm-co.github.io/agenticos/file-processing/) · [Sync sources](https://vstorm-co.github.io/agenticos/howto/configure-sync-sources/)
### 🎨 Publish results as interactive pages
Agents can publish reports, interactive comparisons and small dashboards as **apps**. Choose who can open them; updates keep the same link and earlier versions remain readable. Public links can expire, require a password or restrict which sites may embed them. [Share an app](https://vstorm-co.github.io/agenticos/artifacts/)
### 📊 Track runs, costs and approvals
<img src="docs/assets/readme/dashboard-annotated.webp" alt="The dashboard with six numbered sections: time range, at a glance, runs over time, outcomes, decisions waiting and recent failures." width="100%">
The dashboard answers six questions for a chosen period: how many runs there were, how many finished, what they cost and how many people used agents; how runs changed over time; what failed, waited for approval or was stopped by a budget; where runs came from; and which agents people actually use.
Customize the **dashboard** around your work. **Activity** lets you inspect runs and tool calls, compare agent versions and export records. **Budgets** per agent and organization are checked before each model request. **Approval policies** make sensitive tools wait for a person, and **routines** repeat work on schedules or events such as a new GitHub issue, a Gmail message or a signed webhook. [Run history, budgets and approvals](https://vstorm-co.github.io/agenticos/governance/) · [Routines](https://vstorm-co.github.io/agenticos/triggers/)
### 👥 Organize teams with roles and groups
**Roles define what people may do. Groups define who you share with.** Use roles such as Builder, Operator, Member and Viewer, then create departments or working groups such as Operations, Engineering, Finance and Research. Share an agent, skill, collection, context file or app with a group in one step, or limit it to chosen departments when you create it. [Departments and groups](https://vstorm-co.github.io/agenticos/departments/)
Bring existing company accounts through **OIDC single sign-on** (Entra ID, Okta, Keycloak and others), **LDAP directory login** or **Kerberos integrated Windows sign-in**. **Directory mappings** connect directory groups to a role and a group at sign-in. [Roles and permissions](https://vstorm-co.github.io/agenticos/permissions/) · [Directory sign-in](https://vstorm-co.github.io/agenticos/directory/)
## 🧭 Find your path
Building on it? Go to [For developers and operators](#-for-developers-and-operators).
<details>
<summary><b>Deciding whether to adopt it</b>: the problem it solves, what it takes, how to start</summary>
<br>
| The question your teams cannot answer today | How AgenticOS answers it |
|---|---|
| Who may use which data? | Roles, department groups and per-resource sharing, with company sign-in |
| What does it cost? | Monthly budgets per agent and organization, checked before each model request |
| Who approved that action? | Sensitive tools wait for a person; every decision is recorded |
| Where does our data go? | You run AgenticOS and choose each model, parser and tool it may reach |
**What it takes:** a host (4 vCPU, 8 GB RAM), someone to operate the deployment, and subject experts who maintain instructions and documents. Costs are model usage, infrastructure, external services and people's time; the software is Apache-2.0, commercial use included.
**How to start:** choose one repeated task and the person who will judge the answers, set it up with the model and access rules you choose, and check it against measures agreed beforehand. [Plan a rollout](https://vstorm-co.github.io/agenticos/rollout/) · [Compare approaches](https://vstorm-co.github.io/agenticos/about/comparison/)
</details>
<details>
<summary><b>Reviewing security</b>: data flows, identity, controls, what stays with your IT</summary>
<br>
<img src="docs/assets/readme/security-layers.webp" alt="Six security layers: vault, sandboxes, apps, audit log, sessions and traffic, data hygiene." width="100%">
Security is layered. Credentials sit in an envelope-encrypted vault. Code runs in isolated sandboxes. Published pages are sandboxed. Every organisation has a hash-chained audit log. Sessions are short-lived and revocable, and rate limits apply. Logs are redacted and data is swept on a retention schedule.
| Outbound destination | Used when | Local alternative |
|---|---|---|
| Model provider | Every agent run | Ollama, vLLM or LM Studio on your hardware |
| Embedding provider | Indexing and searching documents | Local Ollama embedding models |
| LlamaParse | Collections set to that parser | PyMuPDF or LiteParse, both local |
| Web search | Agents with web search enabled | Turn the capability off |
| MCP servers, chat channels | Only those you connect | Self-hosted servers, web chat |
| Logfire tracing | Only when a token is configured | Built-in run history |
- **Vault:** a data key per secret, wrapped per organization and key version; master keys rotate; values are never shown again.
- **Sandboxes:** the API holds no Docker socket; containers get no network unless needed, with CPU, process and time limits; gVisor optional.
- **Audit log:** hash-chained per organization, verifiable and exportable.
- **HIPAA profile:** [`deploy/profiles/hipaa/`](deploy/profiles/hipaa/) and `agenticos cmd doctor --profile hipaa` check a running deployment against the §164.312 technical safeguards. It is a configuration check, not a certification. [What it does not claim](https://vstorm-co.github.io/agenticos/security/#the-hipaa-profile-and-what-it-does-not-claim)
- **Stays with your IT:** disk encryption at rest, egress firewall, MFA through your identity provider (no native MFA, SAML or SCIM), and backups that include the vault key.
[Security and data flows](https://vstorm-co.github.io/agenticos/security/) · [Data protection](https://vstorm-co.github.io/agenticos/data-protection/) · [Secrets](https://vstorm-co.github.io/agenticos/secrets/) · [SECURITY.md](SECURITY.md)
</details>
<details>
<summary><b>Looking for a first task</b>: 30 tutorials, each with a check you can run</summary>
<br>
<img src="docs/assets/readme/first-tasks.webp" alt="30 tutorials grouped into documents, support, research and analysis, automation, content and productivity, engineering and safety." width="100%">
[All tutorials](https://vstorm-co.github.io/agenticos/use-cases/). 24 have a reference run by the maintainers; they are starting points, not customer results.
</details>
## 📦 What ships today
<img src="docs/assets/readme/capabilities.webp" alt="Selected built-in capabilities in six groups: knowledge and memory, web, files code and output, how it works, safety and limits, chat channels." width="100%">
<details>
<summary>All 37 built-in capabilities, as text</summary>
- **Knowledge and memory:** knowledge search with citations, skills, context, memory, past conversations.
- **Web:** web search (DuckDuckGo by default; Tavily, Brave or Exa with a key), reading web pages, a web browser the agent drives and you can take over, and computer use with your permission.
- **Files, code and output:** calculations in Python, a sandbox with files and a shell, a coding agent that hands work to Claude Code or Codex, charts, drawings (diagrams, calculators, 3D scenes), image generation (OpenAI or Google), speech (OpenAI, ElevenLabs or Google), video clips (Google, BytePlus, xAI or Alibaba), command watches, Tables, Workflows and apps.
- **How it works:** subagents, planning, thinking, work while tools run, tool search, date and time, instruction reminders, asking the user, drafts and follow-up suggestions.
- **Safety and limits:** guardrails that redact secrets and personal data, long conversations, media offload, tool output limits.
- **Chat channels:** chat channel lookup for Slack, Telegram and Mattermost bots.
Plus any tool from a connected MCP server, and capabilities your engineers add in typed Python. [Capability reference](https://vstorm-co.github.io/agenticos/reference/capabilities/)
</details>
<details>
<summary>Where agents answer, and which models they use</summary>
<img src="docs/assets/readme/eight-surfaces.webp" alt="Examples of places one agent can answer." width="100%">
<img src="docs/assets/readme/model-providers.webp" alt="28 model providers: hosted, your subscription, your cloud contract, or on your hardware." width="100%">
**Where agents answer:** web chat, a website widget, a hosted page, the HTTP API, a streaming WebSocket, Slack, Teams, Discord, Mattermost and Telegram.
**Model providers:**
- **Hosted (22):** OpenAI, Anthropic, Google Gemini, OpenRouter, Mistral, DeepSeek, xAI, Cohere, Groq, Cerebras, Together, Fireworks, Hugging Face, GitHub Models, Alibaba, Moonshot, Z.AI, Nebius, OVHcloud, SambaNova, Heroku and Vercel AI Gateway.
- **Your subscription:** ChatGPT, through Codex, after one sign-in.
- **Through your cloud contract:** Azure OpenAI, AWS Bedrock and Google Vertex AI.
- **Self-hosted:** Ollama and LiteLLM, plus vLLM and LM Studio through an OpenAI-compatible endpoint.
</details>
## 🆚 How it compares
| | Where it runs | Source | Models |
|---|---|---|---|
| **AgenticOS** | **Your infrastructure** | **Apache-2.0** | **28 providers, including local ones and your ChatGPT plan** |
| OpenAI dots | OpenAI's cloud | Proprietary | OpenAI only |
| Grok Bot | xAI's cloud | Proprietary | Grok only |
| Viktor | Viktor's cloud | Proprietary | OpenAI, Anthropic, Google, Kimi |
| n8n | Self-hosted or n8n Cloud | Sustainable Use License | Many, including Ollama |
AI teammates are arriving in every vendor's cloud. AgenticOS is the one you run yourself, on any model, with the source open. [Every comparison, in detail](https://vstorm-co.github.io/agenticos/about/comparison/).
## 🎯 Is AgenticOS the right fit?
Choose it when a team has repeated document or tool-based work, subject experts who can maintain the instructions, and someone responsible for operating a self-hosted deployment. Know where it stops today:
Source-system permissions are not mirrored automatically; an agent can lend its bound knowledge to its users. Self-hosting also requires an operator, and model quality, tool configuration and workload determine results. Read the current [limits and tradeoffs](https://vstorm-co.github.io/agenticos/about/comparison/#what-agenticos-does-not-do-yet).
## 🔐 Own your deployment, models and access
**Sovereign means control over deployment, model providers, data flows and agent access.** AgenticOS is Apache-2.0 software you can inspect, modify and operate.
<img src="docs/assets/readme/sovereignty.webp" alt="Two deployment options: AgenticOS self-hosted with hosted models under your own contract, or fully local with open models through Ollama or vLLM." width="100%">
There are two common setups. **Self-hosted, with hosted models:** AgenticOS, documents, vectors and logs run on your servers, and models come from a provider under your own contract and keys. **Fully local:** the same layer with open models through Ollama or vLLM on your hardware, plus local parsers and tools. Self-hosting the console does not make every model, parser or tool local, so review each destination you configure. [Configure models](https://vstorm-co.github.io/agenticos/models/) · [Security and data flows](https://vstorm-co.github.io/agenticos/security/)
## 🛠️ For developers and operators
AgenticOS is built with FastAPI, Pydantic AI, PostgreSQL with pgvector, Redis and Next.js. Engineers add capabilities, sync connectors and MCP catalog entries in typed Python; teams compose agents from the registered capabilities in the console.
| Layer | What runs there |
|---|---|
| Console | Next.js |
| API | FastAPI |
| Agent runtime | [Pydantic AI](https://ai.pydantic.dev) and [pydantic-ai-harness](https://pydantic.dev/docs/ai/harness/), one runner behind every surface |
| Background work | A job queue in PostgreSQL, served by `python -m app.worker` |
| Data | PostgreSQL with pgvector |
| Code execution | Containers started by `sandboxd` |
Call a published agent with `POST /api/v1/agents/{id}/run` as an authenticated member, or stream tokens over the WebSocket.
[Architecture](https://vstorm-co.github.io/agenticos/architecture/) · [API](https://vstorm-co.github.io/agenticos/api/) · [Add a capability](https://vstorm-co.github.io/agenticos/howto/add-capability/) · [Capability reference](https://vstorm-co.github.io/agenticos/reference/capabilities/) · [Contributing](https://vstorm-co.github.io/agenticos/help/)
The [operating-system analogy](https://vstorm-co.github.io/agenticos/about/) explains the architecture. The optional [desktop app](https://vstorm-co.github.io/agenticos/desktop/) adds Quick Ask, approved local files and computer access, and local Claude Code or Codex. The [mobile guide](https://vstorm-co.github.io/agenticos/mobile/) covers the home-screen console and native development builds.
## ❓ Frequently asked questions
<details>
<summary><b>Is AgenticOS free for commercial use?</b></summary>
Yes. AgenticOS is licensed under Apache-2.0, which allows commercial use, modification and private deployment. You pay for the infrastructure you run it on and for the model providers and external services you choose. The AgenticOS name and logo are not covered by the licence.
</details>
<details>
<summary><b>Can AgenticOS run with local models only?</b></summary>
Yes. Configure Ollama, LiteLLM or an OpenAI-compatible server such as vLLM or LM Studio as the model provider, use local embedding models through Ollama, and parse documents with PyMuPDF or LiteParse. Self-hosting the console alone does not make every model, parser or tool local: check each destination you configure. [Models](https://vstorm-co.github.io/agenticos/models/) · [Data flows](https://vstorm-co.github.io/agenticos/security/)
</details>
<details>
<summary><b>Which sign-in methods and roles does it support?</b></summary>
Email and password with magic links, Google, generic OIDC single sign-on (Entra ID, Okta, Keycloak, Auth0, Authentik, Google Workspace), LDAP and Kerberos. Six built-in roles (Owner, Admin, Builder, Operator, Member, Viewer) combine with department groups and per-resource sharing. Multi-factor authentication comes from your identity provider; there is no native MFA, SAML or SCIM yet. [Permissions](https://vstorm-co.github.io/agenticos/permissions/)
</details>
<details>
<summary><b>How does AgenticOS keep agents under control?</b></summary>
Monthly budgets per agent and per organisation are checked before each model request. Sensitive tools wait for a person's approval. Optional guardrails redact secrets and personal data, and every run is recorded with its agent version, tools, tokens and cost. [Governance](https://vstorm-co.github.io/agenticos/governance/)
</details>
<details>
<summary><b>How is it different from ChatGPT Enterprise, Copilot Studio or n8n?</b></summary>
AgenticOS runs on your infrastructure with any of 28 model providers and has no seat or credit fee of its own. It builds agents for the organisation, published to chat, websites and APIs, rather than assistant seats for individual employees. Compared with n8n it starts from the agent rather than a workflow canvas, and many teams use both. The guides also cover the self-hosted builder [Dify](https://vstorm-co.github.io/agenticos/about/dify/) and coding agents such as Claude Code. [Comparisons](https://vstorm-co.github.io/agenticos/about/comparison/)
</details>
<details>
<summary><b>What does it take to run?</b></summary>
Docker Compose on one host. A machine with 4 vCPU and 8 GB of RAM runs it, and two API workers suit a team of ten. Someone needs to own updates, backups (including the vault key), access and the external services you connect. [Deploy](https://vstorm-co.github.io/agenticos/deploy/) · [Rollout](https://vstorm-co.github.io/agenticos/rollout/)
</details>
## 💬 Community
- **Questions and ideas:** [GitHub Discussions](https://github.com/vstorm-co/agenticos/discussions).
- **Bugs and requests:** [issues](https://github.com/vstorm-co/agenticos/issues); planned work is grouped into [milestones](https://github.com/vstorm-co/agenticos/milestones).
- **Contributing:** read the [contributing guide](CONTRIBUTING.md) and [code of conduct](CODE_OF_CONDUCT.md). Report vulnerabilities privately as [SECURITY.md](SECURITY.md) describes.
- **Releases:** read the [release notes](https://vstorm-co.github.io/agenticos/release-notes/), or choose **Watch → Custom → Releases** on GitHub to be notified.
## 📄 License
[Apache License 2.0](LICENSE). See [NOTICE](NOTICE) and [third-party notices](THIRD_PARTY_NOTICES.md) for attribution and bundled components.
## 🤝 Need help putting agents into production?
Vstorm deploys AgenticOS in client infrastructure, writes the documentation, defines the processes and builds custom capabilities. Maintenance and support are agreed per engagement.
Built with care by [**Vstorm**](https://vstorm.co) · [Vstorm on GitHub](https://github.com/vstorm-co)
## 👥 Contributors
Everyone who has helped build AgenticOS:
<a href="https://github.com/vstorm-co/agenticos/graphs/contributors"><img src="https://contrib.rocks/image?repo=vstorm-co/agenticos" alt="Avatars of everyone who has contributed to AgenticOS"></a>