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

# MCP (Model Context Protocol)

> Connect AI to external tools, data sources, and services with MCP servers

## What is MCP?

[Model Context Protocol (MCP)](https://modelcontextprotocol.io) is an open standard that lets AI models use external tools and data sources. Each MCP server exposes tools the AI can call -- read files, query databases, send messages, and more.

EnConvo has built-in MCP support. Add an MCP server and the AI can use its tools automatically.

## Setting Up MCP Servers

### From Settings UI

1. Open **Settings** → **MCP Servers**
2. Click **Add Server**
3. Choose transport type (stdio or HTTP), fill in the configuration
4. Toggle the switch to enable/disable

### Via CLI

```bash theme={null}
# Add a stdio server
enconvo mcp add github --command npx --args "-y @modelcontextprotocol/server-github"

# Add an HTTP server
enconvo mcp add notion --transport http --url "https://mcp.notion.com/mcp"

# List servers
enconvo mcp servers list

# List tools
enconvo mcp tools list
```

### Import from Other Tools

If you have MCP configs from Claude Desktop, Cursor, or other tools:

1. In **Settings** → **MCP Servers**, click **Import**
2. Paste your JSON configuration
3. Click **Import** to add all servers

### Install Links

MCP servers can be installed with EnConvo deeplinks. It uses the same format as `mcp.json` with a name and transport configuration.

Install links:

```
https://enconvo.com/mcp/install?name=$NAME&config=$BASE64_ENCODED_CONFIG
```

| Component             | Description                                           |
| --------------------- | ----------------------------------------------------- |
| `https://enconvo.com` | Web install page                                      |
| `/mcp/install`        | Path                                                  |
| `name`                | Query parameter for server name                       |
| `config`              | Query parameter for base64 encoded JSON configuration |

The web page auto-launches the `enconvo://` deeplink and shows the server configuration as fallback.

#### Generate install link

1. Get name and JSON configuration of server
2. `JSON.stringify` the configuration then base64 encode it
3. Replace `$NAME` and `$BASE64_ENCODED_CONFIG` with the name and encoded config

#### Example

Try this JSON in the MCP install link generator:

```json Single MCP server config theme={null}
{
  "exa": {
    "url": "https://mcp.exa.ai/mcp"
  }
}
```

Result:

| Format       | Example                                                                                                                       |
| ------------ | ----------------------------------------------------------------------------------------------------------------------------- |
| Text link    | `https://enconvo.com/mcp/install?name=exa&config=eyJ1cmwiOiJodHRwczovL21jcC5leGEuYWkvbWNwIn0=`                                |
| Dark button  | <img src="https://file.enconvo.com/badges/mcp-install-dark.svg?v=20260628-adaptive-logo" alt="Add to Enconvo" height="44" />  |
| Light button | <img src="https://file.enconvo.com/badges/mcp-install-light.svg?v=20260628-adaptive-logo" alt="Add to Enconvo" height="44" /> |

Badge code for your README:

<CodeGroup>
  ```markdown Markdown theme={null}
  [![Add to Enconvo](https://file.enconvo.com/badges/mcp-install-dark.svg?v=20260628-adaptive-logo)](https://enconvo.com/mcp/install?name=your-server&config=BASE64_CONFIG)
  ```

  ```html HTML theme={null}
  <a href="https://enconvo.com/mcp/install?name=your-server&config=BASE64_CONFIG">
    <img src="https://file.enconvo.com/badges/mcp-install-dark.svg?v=20260628-adaptive-logo" alt="Add to Enconvo" height="44" />
  </a>
  ```

  ```jsx JSX theme={null}
  <a href="https://enconvo.com/mcp/install?name=your-server&config=BASE64_CONFIG">
    <img src="https://file.enconvo.com/badges/mcp-install-dark.svg?v=20260628-adaptive-logo" alt="Add to Enconvo" height={44} />
  </a>
  ```
</CodeGroup>

#### Install server

1. Click the link or paste into browser
2. EnConvo opens the MCP Servers settings pane
3. The server is added and starts connecting automatically
4. Once connected, its tools are available to the AI

## Troubleshooting

<AccordionGroup>
  <Accordion title="Server fails to start">
    * Verify the command and arguments in your configuration
    * For stdio servers, ensure the binary is installed and in your PATH
    * For Node.js servers, ensure `npx` or `node` is available
  </Accordion>

  <Accordion title="Tools not appearing">
    * Confirm the server is in **Connected** status
    * Some servers require API keys before exposing tools
    * Try disconnecting and reconnecting the server
  </Accordion>

  <Accordion title="Authentication errors">
    * Verify API keys are correct and not expired
    * Check that required environment variables are set
  </Accordion>
</AccordionGroup>

<Info>
  For the full MCP specification, visit [modelcontextprotocol.io](https://modelcontextprotocol.io).
</Info>