MCP App Store
by KeenEthics
Tudu icon

Tudu MCP Server for Claude and ChatGPT

By Tudu Next, Inc.

Overview

Tudu is where you keep the places and things you want to come back to: restaurants, bars, cafes, hotels, things to do, events, products, recipes and more. This connector brings your Tudu saves into Claude. What you can do: - Search your saves by name, type or city ("the sushi spot I saved in New York", "restaurants I saved in Miami") - Pull up the details of a saved place, including address and booking or website links - Save something new straight from the conversation ("save Joe's Stone Crab in Miami Beach to Tudu") - See your lists, create a new list, and add saved places to a list with a note ("add it to Date night with the note: try the omakase") Great for trip planning, date nights, travel itineraries, restaurant wishlists, bucket lists and gift ideas. Ask Claude to plan a weekend from the places you already saved instead of starting from scratch. Tudu only reads and adds to your own account, under your own sign-in. It never deletes saves or lists, and you can disconnect it anytime from Tudu's connected apps page. Requires a free Tudu account, which you can create when you connect.

How to connect

  1. Claude: Open the Tudu listing in the Claude connectors directory and choose Connect. Claude listing Tudu exposes 6 tools here.
  2. ChatGPT: Open the Tudu listing in the ChatGPT app directory and choose Connect. ChatGPT listing Tudu exposes 6 tools here.

Tools

add_item_to_list

ChatGPT
Add something the user has already saved (a card_id from search_saved_content) to one of their lists (a list_id from list_lists or create_list), with an optional note for that list, such as "try the omakase" or "book 2 weeks ahead". Takes effect immediately. The note belongs to this list only. To add something not saved yet, use save_content with list_id instead.

create_list

ChatGPT
Create a new, private list in the signed-in user's Tudu. The list starts empty; nothing has to be saved first. Check list_lists first so you don't create a duplicate of a list they already have. Returns the new list_id for add_item_to_list.

get_saved_item

ChatGPT
Return one saved item by its card_id, if it is in the signed-in user's Tudu.

list_lists

ChatGPT
Return the lists the signed-in user owns, with item counts and visibility.

save_content

ChatGPT
Save one or more things to the signed-in user's Tudu. Each item is a short line of free text naming the thing, for example "noma sushi NYC" or "Griffith observatory". Include a city and a category when you already know them — places share names, and the city is often the only thing that separates two of them — but never invent either: leave out anything you are not sure of. Tudu resolves each item and builds the card itself, so do not paste descriptions, summaries or URLs you have not been given. Tudu builds each card in the background; this call waits up to about 25 seconds for that and then reports each item as saved (with the card's name, address, category and lists), failed (with the reason), or still being added. Tell the user exactly that, item by item. Only for things NOT already saved: to put an existing save into a list use add_item_to_list, and to make a list (empty or not) use create_list. Saving something again creates a second copy.

search_saved_content

ChatGPT
Search the signed-in user's saved Tudu items (restaurants, hotels, places, events, products). Each word of the query is matched on its own against name, summary, description, cuisine or category, address, city, region and country, and an item must contain every word: "cookies Miami" finds a cookie shop saved in Miami. Use the few words that name what the user wants (a dish, a name, a city), not a sentence. If nothing contains every word, the closest saves come back with match = "some_terms", the thing asked for ranked before the place, and each item lists its missing_words. When the query names a place the user has saves in, a near miss elsewhere carries distance_miles from the middle of those saves, and anything over 30 miles away is left out, so Coral Gables can be offered for Miami and Memphis never is. Never answer "nothing found" when items came back: offer the near miss with its real location, e.g. "No cookie spot saved in Miami, but you saved LIGER'S! Cookies in Coral Gables, about 6 miles away." The same applies when you search one word and filter the results yourself. Returns only what this user can see.

add_item_to_list

Claude

create_list

Claude

get_saved_item

Claude

list_lists

Claude

save_content

Claude

search_saved_content

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/tudu"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/tudu)