AI assistants

Use Boxlist in Claude, ChatGPT and other AI assistants

Boxlist scores films, TV shows, books and games on the same ten taste dimensions. Add Boxlist to your AI assistant and it can use those scores to suggest titles that feel like the ones you love, explain what a title is like, or find something two people will both enjoy. It is free and needs no account.

The server URL

Every assistant below needs this one address. Boxlist uses the Model Context Protocol (MCP), the open standard for connecting assistants to tools.

https://mcp.boxlist.com/mcp

Add it to your assistant

Claude (claude.ai and the desktop app)

  1. Open Customize, then Connectors.
  2. Choose Add custom connector.
  3. Name it Boxlist and paste the server URL. Leave the advanced settings empty: no sign-in is needed.
  4. Choose Add. In a chat, open the tools menu and make sure Boxlist is switched on.

On Team and Enterprise plans an owner may need to add the connector for the workspace first.

ChatGPT

  1. Go to Plugins, select the plus button and choose Add custom MCP server.
  2. Name it Boxlist and paste the server URL under Connection.
  3. Set authentication to No authentication, tick the risk acknowledgement and choose Create as a plugin.
  4. Install the plugin, start a new chat, type @ and pick Boxlist.

If your account still shows the older layout, turn on Developer mode under Settings, Apps, Advanced settings, then use Create to add the same URL.

Claude Code

  1. Run this once in your terminal, then ask for picks in any session.
claude mcp add --transport http boxlist https://mcp.boxlist.com/mcp

Other MCP clients (Cursor, VS Code, Goose and others)

  1. Add a remote server using Streamable HTTP with the server URL. No API key or sign-in.
  2. Clients that use a JSON config file can use the snippet below.
{
  "mcpServers": {
    "boxlist": { "url": "https://mcp.boxlist.com/mcp" }
  }
}

What you can ask

Your assistant picks the right Boxlist tool for you. Each answer links to the title pages on Boxlist.

Similar in feelfind_similar_in_feel
Titles that feel like one you loved, matched on taste dimensions and story themes rather than genre alone. Works across films, TV, books and games.
  • “What should I watch tonight if I loved Succession?”
  • “A book for someone who loved Normal People”
Match for twomatch_for_two
Picks two people will both enjoy, from one to five titles each. Skips anything either of you is likely to dislike.
  • “Something my partner and I will both like: she loves Bridgerton, I love Slow Horses”
Quick taste readquick_taste_read
Turns two to five favourites into a short Taste Fingerprint™ reading, with suggestions and a reason for each.
  • “I love Hades, Disco Elysium and Outer Wilds. What does that say about my taste?”
Title taste profileget_title_taste_profile
How a title feels in plain words, its themes, and where to watch it in the UK and US.
  • “Is Severance worth watching? What is it like?”
  • “Where can I stream Oppenheimer in the UK?”
Search titlessearch_titles
Finds the right film, show, book or game by name, so the other tools pick the version you mean.
  • “Find the Dune book, not the film”

Privacy and terms

  • Free and read-only. There is no account and no sign-in.
  • Each answer has at most five titles, each with a link to its Boxlist page.
  • Nothing you ask is stored. The server keeps anonymous counts of which tool ran and which app called it, never your questions or IP address.
  • Results credit Boxlist with a link. Bulk copying and resale are not allowed.

Read the privacy policy, the MCP server terms and the site terms. Building a product on Boxlist data? See the Recommendation API.

Updated