generated: '2026-09-03'
method: searched
source: https://dialnexa.com/docs/sdks/overview.md + https://dialnexa.com/docs/agents/testing-agents.md + https://dialnexa.com/docs/mcp-tools/quickstart.md
description: >-
  DialNexa has no separate sandbox environment, test-mode key prefixes, or simulated telephony -
  testing happens against production with real (billable) calls. What the provider does publish:
  an interactive playground on every endpoint reference page ("Every endpoint in the reference
  includes a built-in playground for quick testing"), a documented agent-testing flow using phone
  test calls against published agent versions with country validation and dynamic variables, and
  explicit guidance to use a dedicated test workspace when evaluating MCP tools that change data
  or spend wallet balance. The MCP preview_batch_call_leads tool previews leads before a campaign
  is created.
console:
  type: docs-playground
  url: https://dialnexa.com/docs/api-reference/introduction
  note: Mintlify API playground embedded in each endpoint page; runs against production with a real API key.
test_mode:
  separate_environment: false
  test_key_prefixes: null
  test_values: null
  note: No published test phone numbers or fixture values; phone test calls are real calls to a validated recipient number.
practices:
- Use a dedicated test workspace for evaluating state-changing or billable MCP tools.
- Test agents with phone test calls against published versions before assigning live routes.
- Preview batch leads (MCP preview_batch_call_leads) before creating a campaign.