name: Vettly Sandbox and Testing
generated: 2026-09-07
method: searched
source: https://api.vettly.dev/docs
note: >-
  Vettly publishes three testing surfaces: test-mode API keys, a dry-run endpoint,
  and a hosted playground. The OpenAPI security scheme description states keys start
  with vettly_live_ or vettly_test_ (test-mode key prefix documented in the contract
  itself). POST /v1/check/dry-run evaluates a policy against caller-supplied
  mockScores without calling AI providers (also exposed as dryRun() in @vettly/sdk).
  A hosted playground exists at https://vettly.dev/playground (HTTP 200). No test
  fixtures, test clocks, or time simulation are documented.
keys:
  live_prefix: vettly_live_
  test_prefix: vettly_test_
  evidence: "OpenAPI components.securitySchemes.BearerAuth.description (https://api.vettly.dev/docs)"
dry_run:
  endpoint: POST /v1/check/dry-run
  operationId: dryRunCheck
  description: >-
    Test a policy without calling AI providers; pass mockScores (category -> score)
    and receive the allow/warn/flag/block decision the policy would produce.
  docs: https://docs.vettly.dev/api/rest.html#dry-run
playground:
  url: https://vettly.dev/playground
  status: 200
test_fixtures: none-published
test_clocks: none-published