On this page

Connect Qencode to your AI assistant

The Qencode MCP connector lets your AI assistant run video transcoding and processing tasks using plain language instead of raw API calls. Once connected, you can ask your assistant to start jobs, check job status, and pull results directly from a chat or terminal.

Every assistant connects to the same Qencode MCP server: https://mcp.qencode.com/mcp

Before you begin

You'll need a Qencode account. The login page opens automatically during setup.

Please note

Pick your assistant

Claude — Add Qencode through the connectors menu in the message box.
Claude Code — Connect from the terminal with a single command.
ChatGPT — Add Qencode from the Apps directory, or as a custom app in Developer Mode.
Gemini — Add the server to your Gemini CLI settings.
Cursor — Add the server in Cursor's MCP settings.

Connect Qencode to Claude

Claude Chat

Claude connects to Qencode through the connectors menu in the message box. Once you add it, Qencode stays available across your chats.

Open the connectors menu
In the Claude message box, click the + button, select Add connector, then choose Add custom connector.
Enter the connector details
In the "Add custom connector" window, type the name Qencode and paste this server address: https://mcp.qencode.com/mcp. Leave the Advanced settings empty and click Add.
Connect from the directory
Qencode now appears in your directory under Custom connectors. Click it to open the details, then click Connect.
Sign in and authorize access
Your browser opens the Qencode authorization screen. Sign in to your Qencode account, review the requested permissions, select the project you want Claude to work with, and click Allow.
Confirm the connection
Back in Claude, open the + menu again. You'll see Qencode listed with its toggle switched on. Select Add from Qencode to browse the available tools.

Claude Code

Claude Code connects to Qencode through the terminal. Once you add it, Qencode stays available whenever you use Claude Code.

Open your terminal
Open the terminal application on your computer.

Add the Qencode MCP server
Run the following command:

claude mcp add --transport http qencode https://mcp.qencode.com/mcp

Launch Claude Code
Run the following command:
```
claude
```
Authorize Qencode
When your browser opens, sign in to your Qencode account and authorize access.
Open the MCP menu
Return to Claude Code and run:
```
/mcp
```
Confirm the connection
Confirm that Qencode appears in the connected servers list along with its available tools.

Connect Qencode to ChatGPT

ChatGPT connects to Qencode as an app you add from the Apps directory. Once you connect it, Qencode stays available across your chats.

Turn on Developer mode
Go to Settings → Security and login, then switch Developer mode on. You'll see a warning explaining that custom connectors can run third-party code on your behalf. Review the warning and accept it.
Open the connector page
Go to chatgpt.com/plugins, or open Settings → Plugins. The + button or Add custom connector option now appears.
Enter the Qencode server
Paste the following production server address:
```
https://mcp.qencode.com/mcp
```
Authorize Qencode
Qencode appears under Drafts. The first time you invoke it, the Qencode OAuth window opens. Make sure pop-ups are not blocked, then review and approve access.
Enable Qencode in a chat
Start a new conversation, open the tools menu in the message box, and select Qencode for that chat.

Connect Qencode to Gemini

Gemini connects to Qencode through the Gemini CLI. You will install and authenticate the CLI first, then add the Qencode MCP server to its settings file.

Open your terminal and install the Gemini CLI globally using npm or Homebrew. Skip this step if you already have it set up.
```
# Via npm
npm install -g @google/gemini-cli

# Via Homebrew (macOS / Linux)
brew install gemini-cli
```
Get a Gemini API key. Go to aistudio.google.com/app/apikey, click Create API key, name it, choose a project, then copy the key.
Start the CLI:
```
gemini
```
If prompted to trust the folder, choose Trust folder. At the authentication prompt, select Use Gemini API Key, paste the key, and press Enter. The CLI stores it in your system keychain.
Open your Gemini configuration file in a text editor:
- macOS / Linux: ~/.gemini/settings.json
- Windows: %USERPROFILE%\.gemini\settings.json

Add Qencode inside the mcpServers block, then save the file:

{
  "mcpServers": {
    "qencode": {
      "httpUrl": "https://mcp.qencode.com/mcp",
      "timeout": 30000,
      "trust": false
    }
  }
}

Close and reopen your terminal session, then start Gemini again:
```
gemini
```
Authorize Qencode by running the following inside the Gemini CLI prompt:
```
/mcp auth qencode
```
When asked to continue, choose Yes.
A browser window opens to the Qencode authorization page. Sign in with your Qencode account, review the requested permissions, choose your project from the dropdown, then click Allow. The browser confirms authentication and you can return to the terminal.
Confirm the connection by running:
```
/mcp
```
Qencode should appear in the list with a ready status, alongside the video processing tools it provides.

Connect Qencode to Cursor

Cursor connects to Qencode through its MCP settings. You can add the server from Cursor's settings screen, or by editing your configuration file directly.

In Cursor, open Settings, go to Tools & MCP, and click New MCP Server.

Cursor opens an mcp.json file. Add Qencode inside the mcpServers block, then save the file:

{
  "mcpServers": {
    "qencode": {
      "url": "https://mcp.qencode.com/mcp"
    }
  }
}

Sign in to your Qencode account and approve access when your browser opens.
Quit Cursor completely and reopen it. MCP servers load at startup, so a full restart is required.
Return to Tools & MCP and confirm that Qencode shows a green dot and lists the tools it provides.

Need More Help?

For additional information, tutorials, or support, visit the Qencode Documentation page Link or contact Qencode Support at support@qencode.com.