Skip to main content
The Tella MCP (Model Context Protocol) server at https://api.tella.com/mcp connects AI assistants to your workspace. Connect Claude and ChatGPT, or set up Claude Code, Cursor, Codex, and other clients under Settings → Connections → AI apps.

​
What can you do with the Tella MCP server?

Once connected, AI assistants can:
  • Search videos by title or transcript, and access clips, metadata, captions, thumbnails, and chapters.
  • Upload media and create or edit videos, including cuts, layouts, zooms, backgrounds, overlays, sound effects, and Studio Sound.
  • Organize videos with tags, playlists, and playlist groups.
  • Start exports, manage webhooks, and update sharing or presentation settings.
  • Publish videos to your connected YouTube, LinkedIn, X, Instagram and Facebook accounts. See Publish to social media with the API or MCP.

​
Connect an AI assistant to Tella

Click your workspace name in the top-left corner, open Settings → Connections under Account, and scroll to AI apps:
  • For ChatGPT or Claude, click Connect to open Tella in the app’s directory, then sign in and authorize access.
  • For Claude Code, Cursor, Codex, or Other AI tools, click Set up to reveal the command, configuration, or server URL to copy into your client. Sign in to Tella when prompted.
You can also open Tella in Claude’s Connector Directory or Tella in ChatGPT directly. For Muse and Grok Bot, follow the steps below using the same MCP server URL. In ChatGPT, a custom connector pointed at the URL also works, but the official app is simpler. For full setup instructions, configuration options, and available commands, see our MCP documentation: tella.com/docs/mcp-server You can also pick up a skill from the Tella Skills directory: open-source, ready-made workflows, built and tested for Claude Code, that teach your AI agent how to edit videos on local files and publish your latest Tella recording to YouTube. Our agent skills guide walks through installing each one, with example prompts.

​
Connect Tella to Muse

Muse, Meta’s personal AI assistant, reaches apps through Connectors. Tella isn’t in Muse’s connector directory yet, but Muse can build a custom connector for any hosted MCP server, and Tella’s is hosted: Muse connects to https://api.tella.com/mcp from its own cloud environment, on the web, iOS, Android, or WhatsApp. Nothing installs on your machine.
1

Ask Muse to add Tella as a custom connector

Muse has no “add MCP server” setting; you connect by asking. Paste this into a new message:
2

Approve the sign-in link

Muse replies with a Tella sign-in link. Open it, sign in, pick the workspace Muse should use, and authorize access. Muse then discovers the Tella tools, tests the connection, and saves a skill so the connector is ready in later conversations without repeating the setup.
Once connected, ask Muse in plain language: “Find my latest demo video and make it public” or “Trim the first 5 seconds off the intro clip of my onboarding video”. By default Muse asks for your approval before taking actions on your behalf; you can change when it asks under Settings in Muse, and disconnect Tella at any time under Settings, then Connectors.
If Muse asks for an API key instead of a sign-in link, create one in Tella under Settings, then API Keys, and enter it through Muse’s secure credential prompt rather than typing it into the chat. Tella’s MCP server accepts an API key as a bearer token, but the OAuth sign-in above is the simpler route.

​
Connect Tella to Grok Bot

Grok Bot, xAI’s always-on agent, runs each Bot on its own cloud computer and reaches apps through Plugins. Tella isn’t in the plugin marketplace yet, but a Bot can add any hosted MCP server as a custom server, and Tella’s is hosted: the Bot connects to https://api.tella.com/mcp directly, and nothing installs on your machine.
1

Ask your Bot to add Tella as a custom server

Grok Bot has no settings field for an MCP server URL; you add one by asking in chat:
Say “custom MCP server” so the Bot registers the URL instead of searching the marketplace. It repeats the name and URL and asks you to confirm; select Add it.
2

Authorize the connection

The Bot posts a connect card in the chat. Select Authorize, sign in to Tella in the browser, pick the workspace the Bot should use, and allow access. If the card shows Waiting for authorization, select Reopen to bring the sign-in tab back. The Tella tools appear once sign-in completes.
3

Attach Tella to a task

In a chat, type @, pick tella, and ask in plain language: “List my Tella videos” or “Trim the first 5 seconds off the intro clip of my onboarding video”. Bots ask for your approval before changing anything in an external app.
Plugins are account-wide: add Tella once and every Bot on your account can use it. It’s listed under Settings, then Plugins, then Yours, which is also where you disable or remove it. If it shows Disabled by team admin, your team’s MCP policy blocks it; ask an admin to allow https://api.tella.com/mcp.
Grok Bot is not the Grok chatbot. Grok on grok.com and the Grok Build CLI have their own connector settings; the MCP server URL is the same, but the steps above are for the Grok Bot app. If a Bot asks for an API key instead of showing a connect card, create one in Tella under Settings, then API Keys, and hand it over through the Bot’s secrets card rather than pasting it into the chat.

​
Troubleshooting authentication

Connecting to the Tella MCP server runs an OAuth sign-in flow in your browser. When that flow gets stuck (the login never finishes, the AI client keeps reconnecting, or it worked before and suddenly stopped), it’s almost always one of the two issues below.

​
OAuth / callback loop

If clients like Claude Desktop connect through mcp-remote, authorizing opens a browser tab that hands a code back to a local callback. The connection hangs in a loop when that callback can’t complete. Try these in order:
1

Finish the sign-in in your browser

When the AI client starts, a browser tab should open asking you to authorize Tella. Complete that sign-in fully and wait for the “you can close this tab” confirmation before switching back. If no tab opened, your client may have suppressed it. Check the client’s logs for an authorization URL and open it manually.
2

Make sure the local callback isn't blocked

The browser has to redirect back to a local callback URL on your machine. A VPN, proxy, or another process already using the callback port can break the handoff, leaving the client stuck reconnecting. Quit any other AI client that might be running the same MCP connection, then try again.
3

Fully restart the AI client

Quit the client completely (not just close the window) and reopen it so it starts a fresh connection rather than reusing a half-open one.
4

Authenticate the mcp-remote wrapper once from a terminal

If your client uses mcp-remote and restarts it before browser sign-in finishes, run npx mcp-remote https://api.tella.com/mcp in a terminal, complete the sign-in, wait for the confirmation page, then stop the command with Ctrl+C. The wrapper caches credentials in ~/.mcp-auth for the next connection. Codex supports Tella’s HTTP endpoint directly and does not need this workaround; follow the current command in the MCP setup documentation.

​
Reset stale credentials

If a client connected through mcp-remote used to work and is now stuck (for example after changing your password, switching workspaces, or a token expiring), the wrapper’s cached credentials may need to be cleared so the next connection re-runs the full sign-in. The mcp-remote credential cache lives in your home directory at ~/.mcp-auth. To force a clean re-authentication on the next connection, delete that folder: run rm -rf ~/.mcp-auth in a terminal, or remove it in your file manager. Do this only for connections that use the wrapper; native HTTP clients manage credentials themselves.
~/.mcp-auth is in your home directory (the ~ matters). It is not /.mcp-auth at the filesystem root.
After clearing the cache, fully restart your AI client and complete the browser sign-in again.

​
What is MCP?

Model Context Protocol is an open standard that allows AI assistants to securely connect to external tools and data sources. It was created by Anthropic and is supported by Claude and other AI applications.
Last modified on September 30, 2026