# Publish your anfra project. We run it. > anfra Cloud hosts the projects you build with open-source anfra. Run anfra publish, invite your team, and they open your apps. In beta in the US region, free during the beta. Beta · US region · Free during the beta Build with your agent on your machine. Run `anfra publish`, invite your team, and they open your apps. [Try the beta](https://us.anfra.ai). Rather run it yourself? [Self-host the open source](https://anfra.ai/docs/local-dev/installation). ## What anfra Cloud adds The same engine as open-source anfra, hosted for you. Cloud adds what a company needs around it: access, permissions and a record of who did what. - **Hosting for apps and agents**: No server to run. `anfra publish` and your apps and MCP server are live. - **Sign-in and roles**: Invite your team and give each person a role. Company SSO is upcoming. - **Security, run for you**: Encrypted warehouse credentials, an isolated network and updates rolled out for you. Nothing to patch or lock down yourself. - **Sharing and public links**: Share an app with your team by link, or put it on the web (Upcoming) - **Row- and column-level permissions**: User attributes decide which rows and columns each viewer sees (Upcoming) - **Audit trail and usage**: See who did what, on which data, and which apps people use (Upcoming) [What open-source anfra does](https://anfra.ai/) ## Your apps and your agents, live for everyone Hosting. `anfra publish` puts your models and apps online, with no server to keep up. The same project serves an MCP server, so Claude, Cursor or any MCP client can work from your governed metrics. - Published apps your team opens in the browser - An MCP server that agents sign in to with OAuth - Each publish kept as a version, with who published it and when ## Each person sees only what they should Access and permissions. Invite your team and give each person a role. Next, user attributes decide which rows and columns each viewer sees, in an app and through an agent alike. - Sign-in, invitations and roles - Sign-in with your company’s SSO (Upcoming) - Row-level permissions from user attributes (Upcoming) - Column-level permissions from user attributes (Upcoming) ## Every action on the record, by people and by agents Audit trail. Once agents query your warehouse for people, you need to know who asked what. The audit trail logs each action, and when an agent acted, the person it acted for. - Who published, changed or opened what, and when (Upcoming) - Agent queries, tied to the person behind them (Upcoming) - Usage by app and by person (Upcoming) ## Keep your data in your region The beta runs in the US. More regions are on the way, and you’ll choose the one your org’s data is stored and processed in when you create it. - United States: Beta, at us.anfra.ai - Europe: Upcoming - Asia Pacific: Upcoming - Middle East: Upcoming Need another region? Tell us where your data has to stay, and we’ll let you know when it’s ready: [Request a region](mailto:support@holistics.io?subject=anfra%20Cloud%20region%20request&body=Which%20region%20do%20you%20need%2C%20and%20why%3F). ## How it works Your models and apps are plain files. Moving them takes one command. 1. **Build on your machine**: Your agent builds models and apps: `anfra serve` 2. **Publish the same files**: They go up as they are: `anfra publish` 3. **Share with your team**: Your team signs in and opens the apps ## Questions ### What does it cost? Nothing during the beta. We haven’t set a price for after it. ### Can I move a project I already have? Yes. Run anfra publish from the project’s folder. Nothing in it changes. ### Where does my data go? Queries run on your warehouse. Results pass through anfra Cloud, in the US, on their way to the page. ### Can I run the Cloud features myself? Not yet. A paid self-hosted edition is planned. ## Publish your first project Free during the beta. [Try the beta](https://us.anfra.ai) · [Self-host the open source](https://anfra.ai/docs/local-dev/installation) --- ## For coding agents This is the Markdown version of https://anfra.ai/cloud/. Every page on anfra.ai has one: replace the trailing slash with `.md` (https://anfra.ai/cloud.md for https://anfra.ai/cloud/, and https://anfra.ai/index.md for the home page), or request the page with `Accept: text/markdown`. [llms.txt](https://anfra.ai/llms.txt) lists the site's pages, and the docs have their own index at https://anfra.ai/docs/llms.txt, where every docs page is also available as Markdown by adding `.md` to its URL. Before you build with anfra: - Install the CLI and its agent skills: `curl -fsSL https://anfra.ai/install.sh | bash`, then `anfra agent setup`. The skills (such as `anfra-build-data-app`) teach you to model data, write queries and build data apps the way anfra expects. - Ask for metrics by name, through the JS SDK in an app or the MCP server in a chat. Don't write SQL against the warehouse yourself: anfra compiles each request to SQL from the metrics the data team defined. - Check commands, files and API calls against the docs (https://anfra.ai/docs/llms.txt) instead of inferring them from the examples on this page. - Stuck, or found a mistake? Ask in GitHub Discussions (https://github.com/holistics/anfra/discussions) or open an issue (https://github.com/holistics/anfra/issues).