> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cogfoundry.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Installation

> Install the LoomLoom CLI and the integration package for your agent.

The official installer installs the LoomLoom CLI and the integration package for one supported agent. These docs use `v0.1.3-internal.1`, so pin that version to use the commands documented here. The default agent is Codex.

<Note>
  Running the installer without `--version` resolves the latest stable release, which may expose a different command set.
</Note>

## macOS and Linux

```bash theme={null}
curl -fsSL https://raw.githubusercontent.com/Cogfoundry-ai/loomloom/main/install.sh | \
  bash -s -- --version v0.1.3-internal.1
```

Pinned installations download the matching release assets directly.

Choose another supported agent with `--agent`:

```bash theme={null}
# Claude Code
curl -fsSL https://raw.githubusercontent.com/Cogfoundry-ai/loomloom/main/install.sh | \
  bash -s -- --version v0.1.3-internal.1 --agent claude

# OpenClaw
curl -fsSL https://raw.githubusercontent.com/Cogfoundry-ai/loomloom/main/install.sh | \
  bash -s -- --version v0.1.3-internal.1 --agent openclaw
```

## Windows PowerShell

```powershell theme={null}
& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Cogfoundry-ai/loomloom/main/install.ps1))) -Version v0.1.3-internal.1
```

To install the package for another agent:

```powershell theme={null}
# Claude Code
& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Cogfoundry-ai/loomloom/main/install.ps1))) -Version v0.1.3-internal.1 -Agent claude

# OpenClaw
& ([scriptblock]::Create((irm https://raw.githubusercontent.com/Cogfoundry-ai/loomloom/main/install.ps1))) -Version v0.1.3-internal.1 -Agent openclaw
```

## Verify the installation

```bash theme={null}
loomloom --version
```

Next, [configure your credentials](/documentation/loomloom/configure-credentials) and run:

```bash theme={null}
loomloom doctor
```

The agent package uses a small core Skill plus focused reference modules for setup, execution, billing, TemplateSpec, Market, local Skills, and CLI error handling. The installer places the complete package for the selected agent.