Experiments Create

• namestring
  Human-readable display name. Defaults to `flag_key` when omitted.
• intentstring

  Required
  The original user request that led to this tool call. Copy it verbatim when available.
• controlobject
• flag_keystring

  Required
  Developer-chosen handle referenced from code, used in evaluation and results calls. Interchangeable with the returned `expt_` id anywhere the API takes an experiment identifier.
• variantsarray
  Treatment arms to create. Users outside every arm form the implicit `control` group. Required unless `feature_flag_only` is `true`.
• bucket_bystring
  Randomization unit, and the only identity the assignment is keyed on — evaluation fails rather than falling back to another. `user` (default) uses `subject[user_id]` for account experiments and the signed-in user for internal experiments; `account` uses `subject[account_id]`, so…
• intent_idstring

  Required
  A UUID for the user turn that caused this call. Generate it once per user message and reuse it for every tool call made to fulfill that message. Never derive it from the intent text.
• account_idstring
  Owning account tag or internal. Required; ownership cannot change.
• hypothesisstring
  Required for full experiments; rejected on feature flags. Structure it as "If we [change] for [cohort], then [measurable behavior] will [increase/decrease], resulting in [business outcome], because [evidence]. Created by [name]." Evidence should be something real — a baseline nu…
• idempotency_keystring
  Caller-chosen stable retry key. Reuse the same key when retrying after a timeout or error. The native API replays completed requests and rejects requests still in flight.
• targeting_rulesarray
  Rules that determine which subjects qualify for the experiment.
• related_resourceobject

No reviewed structured output summary is available.

No verified related use cases are linked yet.

No published Skills explicitly reference this Action yet.