generated: '2026-08-28'
method: searched
source: https://docs.x.com/tools-and-libraries.md
description: >-
  Testing surfaces X publishes for the X API. X operates NO hosted sandbox or test mode:
  there are no test-vs-live key prefixes, no test fixtures, no time simulation, and every
  call against api.x.com is a real, billed call against real accounts. What X does publish
  is a self-hosted mock server, an interactive Postman workspace, and a bot-account facility
  in the live API for automated testing.
hosted_sandbox: false
test_mode: false
key_prefixes:
  test: null
  live: null
  note: >-
    X issues one set of credentials per app (API key/secret, app-only Bearer, OAuth 2.0
    client id/secret). Nothing in the credential distinguishes test from production.
surfaces:
- name: X API Playground
  type: self-hosted mock server
  url: https://github.com/xdevplatform/playground
  http_status: 200
  official: true
  description: >-
    Local mock server for testing X API v2 endpoints without spending real credits. X's own
    Tools & Libraries page frames it exactly that way — cost avoidance, not a provider-side
    sandbox environment.
  runs: locally
- name: Postman public workspace
  type: interactive collection
  url: https://www.postman.com/xapidevelopers/x-api-public-workspace/collection/34902927-2efc5689-99c6-4ab6-8091-996f35c2fd80
  http_status: 200
  official: true
  description: Interactive API testing for all v2 endpoints. Calls go to the live API with live credentials.
- name: Bot accounts
  type: live-API test identity
  operations: [createBot, getBots, updateBot, deleteBot, revokeBotToken]
  description: >-
    POST /2/bots creates a bot account inside the calling app's project and returns a bearer
    token, and the operation is documented as idempotent on handle. It gives an automated
    caller its own identity to act as, but on the production platform, at production prices.
  source: openapi/_original/x-api-v2-openapi.json
- name: Developer Console
  type: console
  url: https://console.x.com
  http_status: 200
  description: Credential management, live usage and cost tracking, spending limits, per-endpoint rate limit tables.
test_values:
  published: false
  note: >-
    X publishes no test cards, test tokens, test accounts or fixture ids. None are invented here.
cost_of_testing:
  note: >-
    Because every live call is billed per resource or per request, and writes are not
    refundable, the practical guardrails X offers are the spending limit and auto-recharge
    settings in the Developer Console plus the 24-hour read deduplication window.