get_assistant
ChatGPTFetch an assistant configuration. Engine and primary recognition model choices require Fallbacks & Guardrails; platform-managed voice models remain private.
get_assistant_version
ChatGPTFetch one version including its configuration snapshot. Model choices require Fallbacks & Guardrails.
get_call
ChatGPTFetch a single call including measured call insights (unavailable values and unavailable insights are null), transcript messages, a ready-to-read structured_transcript ("ai: …" / "human: …" lines, one per turn), typed tool calls/results, chronological events, summary, evaluation, Credits billed (billing.credits_charged), Telephony (billing.telephony_credits after the live carrier price at the workspace rate), credit hold (billing.credits_reserved / credits_released / hold_reason), a provider-neutral failure when applicable, and a time-limited recording_url (valid for 1 hour). Transfer-failure events expose only their safe failure object, never raw connection details. Buy/provider cents are never included.
get_email_history_item
ChatGPTFetch a complete multi-turn email conversation in chronological order, including stable thread and parent IDs, reply statuses, errors, attachment metadata, available inbound audio transcripts and short image descriptions including readable text when possible, with media links valid for one hour and advisory message labels when available.
get_languages
ChatGPTLists the platform's assistant language catalog ({ code, label }, ISO 639-1 or ISO 639-3). Available codes for primary and secondary languages also depend on the assistant's TTS, STT, and realtime models.
get_models
ChatGPTRequires Fallbacks & Guardrails for the active workspace. Lists available model choices filtered by workspace inference region and engine compatibility. Fallback chains are managed automatically. Null follows workspace then platform defaults. Without 'type', results are grouped by type.
get_voices
ChatGPTBrowse the workspace voice library using provider-neutral local selectors. Results can be filtered by language, gender, accent, or free-text search. Each result has a voice ID and mode-specific compatibility. Full Duplex uses its native voice for greetings, consent, and tool announcements.
list_assistant_versions
ChatGPTList saved configuration snapshots for an assistant (newest first).
list_assistants
ChatGPTList all AI phone assistants of the account, including prompt and voice settings. Model choices are returned only with Fallbacks & Guardrails. Supports limit/offset pagination.
list_history
ChatGPTList calls, messaging, and assistant email conversations together, newest first. Failed outbound call rows include a provider-neutral failure object. Email turns with the same stable thread ID are grouped into one conversation. Filter by Audience contact, contact tags, type, direction, status, assistant, campaign, sentiment, outcome, QA result, duration, search text, relative date range or time range — list filters accept one value or a list (any of). Pass view_id to apply a saved History view (list_history_views); explicit filters override the view.
list_prompt_templates
ChatGPTLists prompt and flow templates visible to this workspace (platform and workspace or reseller). Each template has an ID and blueprint with surfaces, outcome, setup, and avatar presentation; compatibility summarizes graph support. The avatar URL is the catalog portrait, not the virtual speaking avatar. Filters: language, theme, branch, q, has_flow. Returns templates and pagination (limit, offset, total); default 50 templates, maximum 100 per page.