Getting Started

On this page

Prerequisites

Before installing Open Poet, make sure you have the following available on your system:

Installation

Homebrew (macOS & Linux)

Terminal 
brew install openpoet/tap/openpoet

Windows

Download the .zip for your architecture from the Downloads page, extract it, and add the binary to your PATH.

Shell Script (macOS & Linux)

Terminal 
curl -fsSL https://raw.githubusercontent.com/openpoet/openpoet/main/install.sh | sh

Manual Download

Download the binary for your platform from the Downloads page or the GitHub Releases.

Build from Source

Requires Go 1.24+:

Terminal 
git clone https://github.com/openpoet/openpoet.git
cd openpoet
make build

The compiled binary will be placed in ./build/openpoet. Add it to your PATH or move it to a directory in your PATH to use it as openpoet.

Upgrade

To upgrade to the latest version, run the install script again:

Terminal 
curl -fsSL https://raw.githubusercontent.com/openpoet/openpoet/main/install.sh | sh

If you installed via Homebrew:

Terminal 
brew upgrade openpoet

Uninstall

To remove Open Poet installed via the shell script:

Terminal 
curl -fsSL https://raw.githubusercontent.com/openpoet/openpoet/main/install.sh | sh -s -- --uninstall

This removes the binary and optionally the data directory (~/.openpoet/) after prompting for confirmation.

Running Open Poet

Start the server:

Terminal 
openpoet

Open http://localhost:8080 in your browser.

CLI Options

Terminal 
openpoet [flags]
openpoet version       Print version
openpoet mcp-serve     Run as MCP server

CLI Flags

Flag Default Description
-bind localhost Address to bind the server to
-port 8080 Port number for the HTTP server
-db ~/.openpoet/openpoet.db Path to the SQLite database file
-openai-key OpenAI API key for voice transcription
-mcp-http false Enable MCP HTTP endpoint at /mcp

First Steps

Once the server is running, follow these steps to set up your first project:

  1. Open the dashboard — navigate to http://localhost:8080 in your browser.
  2. Add your first project — provide a local file path or an SSH remote (e.g. user@host:/path/to/project).
  3. Configure AI provider settings (optional) — go to Settings and enter your Anthropic API key.
  4. Start a Claude Code session — open your project and click New Session to launch an interactive terminal.
  5. Create skills (optional) — define instruction templates that customize how Claude Code behaves in your sessions.

Next Steps

Now that Open Poet is up and running, explore these guides to unlock its full potential: