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

# Config Center

> Manage Agents, MCP servers, and Skills in the Config Center.

**Config Center** is Vibex's configuration entry point, opened from the **Config Center** button in the sidebar or from the system tray. It brings Agent runtimes, MCP servers, and Skills together in one place, and shows status and validation results for each resource.

## Sections

The Config Center has three tabs across the top:

| Tab        | Purpose                                                                                         |
| ---------- | ----------------------------------------------------------------------------------------------- |
| **Agent**  | Add, install, sign in to, and manage Agent runtimes, and configure model providers inside them. |
| **MCP**    | Add MCP servers and enable them per Agent.                                                      |
| **Skills** | Import reusable `SKILL.md` capability packages and enable them per Agent.                       |

<Note>
  **The Config Center currently offers only these three sections.** Model providers are not a separate tab—they live in the details area of the **Agent** tab. See [Agents and provider profiles](/docs/en/agents-and-providers).
</Note>

## Common operations

### Save and validate

1. Fill in the fields and save.
2. Click **Validate** on the resource card to run validation. Validation is **offline**: it checks required fields, format, and scope, but starts no processes and makes no network requests.
3. For model providers you can also run **Test connection**, which makes one real call to the model API.

<Warning>
  **Test connection** and **Fetch models** reach the provider's network and may incur charges. Confirm the address, model, and permissions before you run them.
</Warning>

### Scope

Some resources let you set a scope, normally **global**, **user**, **project**, or **workspace**. Scope determines which sessions and projects can see the resource.

Start with the smallest scope:

* Put resources that only you use across all projects in **user**.
* Put team or repository conventions in **project**.
* Put resources that serve only the current checkout in **workspace**.
* Use **global** only when you need a global default.

<Info>
  The MCP server editor currently offers only the **User** and **Workspace** scopes.
</Info>

### Enable and disable

Resources relate to Agents through **per-Agent enablement**. The **Agent enablement** area below a resource card provides a switch for each added Agent; only Agents with the switch on load the resource.

When you no longer need a resource, **disable** it first instead of deleting it right away—disabling keeps the record so you can restore it later.

## Credentials

Reference credentials through a secret reference or your operating system's secure storage. Do not paste tokens, private keys, full environment variables, prompts, or file contents into configuration descriptions or diagnostic logs.

The desktop stores credentials in the operating system keychain. Hosts without a usable keychain fall back to writing a file that only its owner can read; see [Self-hosted headless runtime](/docs/en/self-hosted-server).

## Export to Agent-native configuration

On a resource card you can use **Preview export** / **Apply** / **Rollback** to write MCP or Skills from Vibex into the matching Agent's own configuration files (for example Codex's `~/.codex/config.toml`, Claude's `~/.claude/settings.json`, and each Agent's Skills directory).

<Warning>
  **Always review the preview before you Apply.** By default Vibex does not rewrite an Agent's native home configuration; it writes only when you explicitly apply. After applying, you can restore from the same place with **Rollback**.
</Warning>

## When configuration goes wrong

1. Confirm the resource's scope and enablement state.
2. Run **Validate** to see whether there are structured errors.
3. Confirm the resource is switched on in the target Agent's **Agent enablement**.

If state is inconsistent after a restart, see [Troubleshooting](/docs/en/troubleshooting). Do not delete the SQLite database directly.