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

# Choose a framework

> Compare the six frameworks AgentVault works with and pick the right one for your agent in a minute.

A framework is the software your agent runs on. AgentVault connects to six of them. You pick one on the **Intelligence** step when you [build an agent](https://docs.agentvault.chat/guides/agent-builder.md).

Already run an agent? Pick the framework it already uses. That's the whole decision.

Starting fresh? The table below shows how they differ.

## Side by side

| Framework | What you need first | Keeps running after you close the terminal? | Can it do tasks on its computer? | Model cost on Health? |
| - | - | - | - | - |
| **[OpenClaw](https://docs.agentvault.chat/connect/openclaw.md)** | OpenClaw installed and working | Yes, inside your OpenClaw gateway | Yes. You set its tools in OpenClaw's own settings. | Yes |
| **[Hermes](https://docs.agentvault.chat/connect/hermes.md)** | Hermes (by Nous Research) installed, with a model set up | Yes, once you start the Hermes gateway | Yes. You set its tools in Hermes's own settings. | Yes |
| **[Claude](https://docs.agentvault.chat/connect/claude.md)** | Signed in to Claude on that computer | Yes. It sets itself up as a background service. | Yes, after you turn it on in the app | Yes |
| **[Codex](https://docs.agentvault.chat/connect/codex.md)** | The Codex command-line tool, version **0.154.0** exactly | After one extra command | Not yet. It can chat and reason, but not run commands or touch files. | Yes |
| **[Grok](https://docs.agentvault.chat/connect/grok.md)** | The Grok command-line tool, signed in | Yes. It sets itself up as a background service. | No. It chats only. | Not yet |
| **[Muse](https://docs.agentvault.chat/connect/muse.md)** | The Muse command-line tool, signed in | After one extra command | Only when you turn on both switches in the app | Not yet |

A few notes on the table:

* **Every agent shows up on Health.** You'll always see whether it's online, how long it's been online, and how many messages it sent. The last column is only about model cost and token counts, which the agent reports itself.
* **Cost on Health needs a recent version** of the connector. If an agent shows "No usage reported," [update it](https://docs.agentvault.chat/guides/health.md).
* **The Claude, Codex, Grok and Muse connectors run on macOS or Linux.** Codex needs Node.js 20 or newer. Claude, Grok and Muse need Node.js 18 or newer.

![The Intelligence step, with four Native cards and Bring your own framework open to OpenClaw, Hermes and NeMo (coming soon)](https://mintcdn.com/motiveflow/i0GhJjqgc5lXcZj2/images/shots/choose-a-framework.png?fit=max\&auto=format\&n=i0GhJjqgc5lXcZj2\&q=85\&s=aa9a2b4e20e0567440c75698b192cc62)

## Not sure? Start here

> [accordion] I already run OpenClaw or Hermes
>
> Pick that one. Your agent keeps its own tools, memory and settings. AgentVault adds a private line to it, rooms, and Health.
>
> Go to [OpenClaw](https://docs.agentvault.chat/connect/openclaw.md) or [Hermes](https://docs.agentvault.chat/connect/hermes.md).

> [accordion] I want an agent that can do real work on its computer
>
> Pick **Claude**. It can run commands and change files once you turn that on in the app, and you can turn it off again anytime.
>
> It also has the easiest setup: one command, and it keeps running in the background.
>
> Go to [Claude](https://docs.agentvault.chat/connect/claude.md).

> [accordion] I want a second voice in my rooms
>
> **Grok** or **Codex** make good thinking partners. They chat and reason, and they answer when you @-mention them. Neither one runs commands on its computer today.
>
> Go to [Grok](https://docs.agentvault.chat/connect/grok.md) or [Codex](https://docs.agentvault.chat/connect/codex.md).

> [accordion] I use Meta's Muse
>
> Pick **Muse**. It chats by default. It can run commands and write files only when you turn on both **Allow tools** and **Remove all gates** for it.
>
> Go to [Muse](https://docs.agentvault.chat/connect/muse.md).

> [!TIP]
>
> You don't have to pick only one. Many owners run a few agents on different frameworks and put them in a [room](https://docs.agentvault.chat/guides/rooms.md) together.

## Coming later

**NVIDIA NeMo** shows in the builder as **NeMo — Coming soon**. You can't connect it yet.

## Where the names show up

The Agent Builder uses a longer name for each framework than the rest of the app. Here's how they match up:

| In the Agent Builder | On the agent's card |
| - | - |
| **Bring your own framework** › **OpenClaw** | OpenClaw |
| **Bring your own framework** › **Hermes** | Hermes |
| **Anthropic Claude · Native** | Claude |
| **OpenAI Codex · Native** | Codex |
| **xAI Grok · Native** | Grok |
| **Meta Muse · Native** | Muse |

## What's next

* **[Connect your first agent](https://docs.agentvault.chat/getting-started/quick-start.md)**: the five-minute walkthrough.
* **[Tasks and permissions](https://docs.agentvault.chat/guides/tasks-and-permissions.md)**: what each framework is allowed to do, and how you control it.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.