# Installation
Bento is a command-line daemon for macOS and Linux. The installer selects the
published binary for the current machine and installs it in user space.
## Supported platforms
M
Automatic
macOS · Apple Silicon
M-series Macs · arm64
bento-darwin-arm64
M
Automatic
macOS · Intel
Intel Macs · x86\_64
bento-darwin-x64
L
Automatic
Linux · x64
64-bit Linux · x86\_64
bento-linux-x64
L
Automatic
Linux · arm64
64-bit Linux · arm64
bento-linux-arm64
Windows is not currently supported.
## Install automatically
The one-line installer supports all platforms listed above. It verifies the
release checksum, then installs the binary under `/usr/local/bin`,
`/opt/homebrew/bin`, or `~/.local/bin`, depending on directory permissions.
```bash
curl -fsSL https://install.getbento.sh | sh
```
:::warning\[Do not use sudo]
The daemon runs in user space and does not need elevated permissions. If the
command is not found after installation, check that `~/.local/bin` is in your
`PATH` and that the binary is executable.
:::
Verify the installation:
```bash
bento --version
```
## Install a specific version
Set `BENTO_RELEASE` for a reproducible install or rollback:
```bash
curl -fsSL https://install.getbento.sh | BENTO_RELEASE=0.6.5 sh
```
## Requirements
* [Docker](https://docs.docker.com/get-docker/) for the Postgres database and to run agent workloads in secure sandboxes
* [OrbStack](https://orbstack.dev/) is a great Docker alternative for Mac with better performance and native UI.
* [Claude Code](https://claude.ai/code) (`claude`) our default agent runtime
* [GitHub CLI](https://cli.github.com/) (`gh`) for GitHub integration
* [QMD - Query Markup Documents](https://github.com/tobilu/qmd) (`qmd`) for knowledge retrieval
```bash
npm install -g @tobilu/qmd
```
* **OSX only**
* SQLite — `brew install sqlite`
## Optional
* [Codex CLI](https://developers.openai.com/codex/cli) (`codex`) for an alternative agent runtime
## Bootstrap a project
Navigate to your project directory and run:
```bash
bento init
```
This creates `.bento/daemon.yaml` (daemon config) and some starter files.
## Setup the database
By default, the daemon looks for Postgres at `postgresql://bento:bento@localhost:8421/bento`.
Editable in `.bento/daemon.yaml`, but for now, let's get a Postgres instance running at that URL. The easiest way is with Docker:
```bash
curl -fsSL https://install.getbento.sh/services.yml -o docker-compose.yml
docker compose up -d
```
`docker compose down -v` removes the volume and all data.
## Start the daemon
It's designed to be noisy if something is wrong, so if it starts without errors, you're good to go:
### Serving in the foreground
```bash
bento serve
```
### Installing as a background service
```bash
bento daemon install # registers launchd (macOS) or systemd (Linux)
bento daemon start
```
Check everything is healthy:
```bash
bento doctor
```
## Your first pipeline
`bento init` scaffolded a `CLAUDE.md` and `.mcp.json` into your repo. Open the repo in an AI coding agent (Claude Code, Codex, Cursor): it reads `CLAUDE.md` to learn how bento works and connects to the `bento` and `bento-docs` MCP servers declared in `.mcp.json`. Ask it to build your first pipeline — agent persona, pipeline YAML, webhook registration, and a test run.
For the concepts behind agents, skills, and pipelines, see [Concepts](/concepts). For the full pipeline field reference, see [Pipelines: Config](/pipelines/config).
## Configuration reference
`.bento/daemon.yaml` controls daemon-wide settings (database URL, ports, sandbox
backend, pipeline lifecycle hooks). Each file under `.bento/pipelines/` is one
pipeline definition. See [Configuration](/configuration) for the full field reference.
**Runtime state** (PID, log, agent dirs) lives under `~/.bento//` where
`name` is the `name:` field in `daemon.yaml` (defaults to `bento`). Workspaces
are shared at `~/.bento/workspaces/`. Neither directory is committed to git.