Skip to main content
Every read returns each row with every public field by default. Add fields to keep only the fields you name. Add omit to drop fields and keep the rest. The price stays the same. A row costs the same with 2 fields or 200.

​
Keep the fields you need

Name fields in a comma-separated list. Use a dot for a nested field.
Cursors, counts, and diagnostics always stay, so pagination keeps working. The row count and row order never change.

​
Drop the fields you do not need

omit takes the same names. It keeps every other field.
Send both to keep a parent and drop part of it. fields=id,author&omit=author.description returns each author without its bio.

​
How to name a field

  • Case and underscores do not matter. likeCount and like_count name 1 field.
  • A row that lacks a field leaves it out. The API adds no empty value.
  • A name may start with the list it belongs to. tweets.id equals id.
  • Each parameter takes up to 100 names.

​
Lookups return 1 record

A lookup keeps the named fields in each record of its answer. GET /x/tweets/{id} returns a tweet and its author.
GET /x/users/{id} returns the profile itself, so name its fields directly. fields=id,username,followers returns those 3 fields.

​
Unknown names answer 400

A name the answer does not document returns 400 unknown_field. Xquik checks names before it reads X, so the request costs nothing. The error names a close field when there is one, and lists the valid ones.
allowedFields holds every valid name at the place the unknown name stopped. The example shows 3 of them.

​
Use it from MCP, RapidAPI, and SDKs

  • MCP read tools take fields and omit with the same names. A smaller answer fits more rows in 1 tool result.
  • RapidAPI answers use snake_case. Name fields in either style. fields=like_count and fields=likeCount return like_count.
  • SDK methods pass both parameters through. A typed row then holds only the fields you named.

​
Which endpoints take it

Every read that returns rows or 1 record takes fields and omit. That covers tweets, replies, quotes, timelines, users, followers, lists, communities, trends, jobs, Spaces, and lookups. Endpoints that return bare IDs, such as follower IDs, ignore both parameters. Each endpoint page in the API reference lists them. For every field a tweet or profile can hold, see the API field guide.