# Installation

> Documentation index: [llms.txt](https://namera.ai/llms.txt).

Source: https://namera.ai/docs/mcp/installation

Install the local MCP server included in the Namera CLI.

## Requirements [#requirements]

Use **Node.js 24.14+**, an accessible OS keyring, and a client that can launch
local stdio MCP servers. Install Namera on the machine where that client runs.
On Linux, use a persistent Secret Service session so saved credentials and
imported keys remain available after reboot.

## Install [#install]

<Tabs items={["npm", "pnpm", "yarn", "bun"]}>
<Tab value="npm">

```sh
npm install -g @namera-ai/cli
```

</Tab>
<Tab value="pnpm">

```sh
pnpm add -g @namera-ai/cli
```

</Tab>
<Tab value="yarn">

```sh
yarn global add @namera-ai/cli
```

</Tab>
<Tab value="bun">

```sh
bun add -g @namera-ai/cli
```

</Tab>
</Tabs>

Use one command. The Yarn command requires Yarn Classic; with modern Yarn, use
npm or pnpm for the global install. Bun installs the package, but the executable
still runs on Node.js.

```sh
namera --version
```

Continue to [Getting Started](https://namera.ai/docs/mcp/getting-started) or go directly to
[your client's setup page](https://namera.ai/docs/mcp/clients).

## Update [#update]

```sh
npm install -g @namera-ai/cli@latest
```

Restart the MCP server in your client after updating. Saved connections and
imported keys are retained.