Die Beschreibungen der Endpoints und der Felder erscheinen auf Englisch, genau so, wie die API sie ausliefert. Das ist Absicht: Sie lesen hier, was Sie auch in den Antworten sehen.
GET /v1/agents#
List agents. Returns all agents on your account.
If your operator has configured guardrails, the expanded agent carries them read-only, and guardrails.input.action_by_channel says what the input guard's action means on each channel **for this agent**. It matters for one value: escalate is a real handover to a person only in a **chat** on an agent whose web_widget.enabled and web_widget.takeover.enabled are both on, because the handover runs through the same path the request-a-human tool uses and that path refuses when either switch is off. On a **voice** call it is refuse plus the fallback line and nothing more, since a phone leg has no takeover queue to enter. So one stored escalate reads back as {"voice": "refuse", "chat": "escalate"} on an agent with the widget handover armed and as {"voice": "refuse", "chat": "refuse"} on one without it - neither is a misconfiguration, they are the same setting reported honestly for the agent you asked about. The field is filled by the server on reads and ignored on writes.
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl "https://api.voiceland.ai/v1/agents" \
-H "Authorization: Bearer VL_API_KEY"
Beispielantwort
{
"items": [
{
"deploy_state": "deployed",
"name": "front-desk"
}
]
}
POST /v1/agents#
Create or update an agent. Upserts an agent from the narrow tenant shape. Server-managed knobs (model, transcriber tuning, durations, tools) are filled from your account defaults; the response returns the fully expanded agent so you can see what was applied. Unknown fields are rejected with 422.
If your operator has configured guardrails, the expanded agent carries them read-only, and guardrails.input.action_by_channel says what the input guard's action means on each channel **for this agent**. It matters for one value: escalate is a real handover to a person only in a **chat** on an agent whose web_widget.enabled and web_widget.takeover.enabled are both on, because the handover runs through the same path the request-a-human tool uses and that path refuses when either switch is off. On a **voice** call it is refuse plus the fallback line and nothing more, since a phone leg has no takeover queue to enter. So one stored escalate reads back as {"voice": "refuse", "chat": "escalate"} on an agent with the widget handover armed and as {"voice": "refuse", "chat": "refuse"} on one without it - neither is a misconfiguration, they are the same setting reported honestly for the agent you asked about. The field is filled by the server on reads and ignored on writes.
Anfragetext
application/json Schema: Agent
| Feld |
Typ |
Erforderlich |
Beschreibung |
calendar_id |
string |
|
|
call_webhook |
CallWebhook |
|
Optional post-call POST: transcript + summary + extracted custom fields + audio URLs, delivered with retry + HMAC signing. |
cdr_export |
CDRExport |
|
Optional lean metadata-only post-call POST for billing / archive systems (no transcript or summary). |
data_sources |
array of string |
|
Ids of data sources (ds-…) whose latest snapshot is placed in the agent's context at call start. Up to 8. |
event_webhooks |
EventWebhooks |
|
Optional real-time per-event delivery (answered / caller utterance / agent utterance / hangup / transfer). |
identity |
Identity |
Ja |
The agent's system prompt and first spoken line, your primary content surface. |
intents |
IntentTaxonomy |
|
|
knowledge_collections |
array of string |
|
|
language |
Language |
Ja |
Primary BCP-47 language (drives speech recognition + the default summary language) plus optional secondary fallbacks. |
name |
string |
Ja |
Logical identifier: lowercase, alphanumeric plus dash/underscore, ≤64 chars. |
notifications |
Notifications |
|
Optional per-call email recap to a list of recipients, independent of call_webhook. |
outbound |
boolean |
|
When true the agent is outbound-only (can be dialed via /calls/dial); when false (default) it is inbound-only and answers calls routed to its numbers. |
pickup_delay_s |
integer |
|
Seconds to wait after answering before the agent speaks its first line (0 to 15). Use 1 to 2 for a more human pace. |
re_engagement_prompt |
string |
|
Line spoken when the caller has been silent for the re-engagement interval. |
record_call |
boolean |
Ja |
When true, per-call stereo + per-channel recordings are produced and their URLs exposed. |
search_widget |
SearchWidget |
|
Optional public knowledge-search box rendered by /search-widget.js on your site. Independent of web_widget: enabling the chat bubble never enables this, and this serves anonymous readers rather than a conversation. |
timezone |
string |
|
IANA timezone for any wall-clock formatting in prompts/tool args. Empty falls back to the account default. |
voice |
Voice |
Ja |
The synthesized voice: provider + voice id (+ optional model). |
web_widget |
WebWidget |
|
Optional browser-embed voice/chat bubble rendered by /embed.js on your site. |
Antworten
| Code |
Beschreibung |
200 |
The expanded agent. |
401 |
Missing or invalid API key. |
402 |
The agent uses a capability your plan does not include (feature_not_entitled, the message lists the feature codes), or the account has no subscription (no_subscription). |
409 |
Creating this agent would exceed your plan's agent allowance, one pool shared by voice and text agents, counted across your whole account (agent_limit, the message says how many the plan allows and how many exist), or this project's own agent limit. Updating an existing agent is never refused for this reason. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X POST "https://api.voiceland.ai/v1/agents" \
-H "Authorization: Bearer VL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"identity": {
"initial_message": "Acme Clinic, how can I help?",
"system_prompt": "You are the front desk of Acme Clinic. Help callers book appointments."
},
"language": {
"primary": "en-US"
},
"name": "front-desk",
"record_call": true,
"voice": {
"provider": "voiceland",
"voice_id": "pNInz6obpgDQGcFmaJgB"
}
}'
Beispielantwort
{
"api_version": "2026-04-26",
"deploy_state": "draft",
"name": "front-desk"
}
POST /v1/agents/generate#
Generate an agent from a brief. Uses the language model to draft a complete agent from a free-text description, then upserts it. Returns the expanded agent.
Anfragetext
application/json Schema: AgentGenerateRequest
| Feld |
Typ |
Erforderlich |
Beschreibung |
description |
string |
Ja |
Free-text description of what the agent should do. |
language |
string |
|
|
name |
string |
Ja |
Name for the generated agent. |
pickup_delay_s |
integer |
|
|
record_call |
boolean |
|
|
timezone |
string |
|
|
voice_id |
string |
|
|
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X POST "https://api.voiceland.ai/v1/agents/generate" \
-H "Authorization: Bearer VL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"description": "A friendly receptionist that greets callers and answers general questions about the business.",
"language": "en-US",
"name": "greeter",
"record_call": true
}'
Beispielantwort
{
"deploy_state": "draft",
"name": "clinic-bot"
}
POST /v1/agents/wizard#
Draft an agent from structured input. Builds an agent from a small set of structured fields (industry, language, voice persona, transfer behavior). Non-persisting, returns the draft yaml + tenant agent for review.
Anfragetext
application/json
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X POST "https://api.voiceland.ai/v1/agents/wizard" \
-H "Authorization: Bearer VL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"industry": "receptionist",
"language": "en-US",
"name": "front-desk",
"voice": {
"gender": "female",
"style": "warm"
}
}'
Beispielantwort
{
"name": "front-desk",
"warnings": [],
"yaml": "name: front-desk\n…"
}
DELETE /v1/agents/{name}#
Delete an agent. Permanently deletes the agent.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
Antworten
| Code |
Beschreibung |
204 |
Deleted. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X DELETE "https://api.voiceland.ai/v1/agents/{name}" \
-H "Authorization: Bearer VL_API_KEY"
GET /v1/agents/{name}#
Get an agent. Returns the fully expanded agent spec, including the server-filled defaults.
If your operator has configured guardrails, the expanded agent carries them read-only, and guardrails.input.action_by_channel says what the input guard's action means on each channel **for this agent**. It matters for one value: escalate is a real handover to a person only in a **chat** on an agent whose web_widget.enabled and web_widget.takeover.enabled are both on, because the handover runs through the same path the request-a-human tool uses and that path refuses when either switch is off. On a **voice** call it is refuse plus the fallback line and nothing more, since a phone leg has no takeover queue to enter. So one stored escalate reads back as {"voice": "refuse", "chat": "escalate"} on an agent with the widget handover armed and as {"voice": "refuse", "chat": "refuse"} on one without it - neither is a misconfiguration, they are the same setting reported honestly for the agent you asked about. The field is filled by the server on reads and ignored on writes.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl "https://api.voiceland.ai/v1/agents/{name}" \
-H "Authorization: Bearer VL_API_KEY"
Beispielantwort
{
"deploy_state": "deployed",
"name": "front-desk",
"spec": {
"guardrails": {
"input": {
"action": "escalate",
"action_by_channel": {
"chat": "escalate",
"voice": "refuse"
},
"enabled": true
}
}
}
}
POST /v1/agents/{name}/deploy#
Deploy an agent. Publishes the agent so it answers/originates live calls.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X POST "https://api.voiceland.ai/v1/agents/{name}/deploy" \
-H "Authorization: Bearer VL_API_KEY"
Beispielantwort
{
"deploy_state": "deployed",
"name": "front-desk"
}
POST /v1/agents/{name}/rollback#
Roll back an agent. Restores a prior version as the current spec.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
Anfragetext
application/json Schema: RollbackRequest
| Feld |
Typ |
Erforderlich |
Beschreibung |
version_id |
string |
Ja |
The version to restore. |
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X POST "https://api.voiceland.ai/v1/agents/{name}/rollback" \
-H "Authorization: Bearer VL_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"version_id": "v_…"
}'
Beispielantwort
{
"deploy_state": "draft",
"name": "front-desk"
}
POST /v1/agents/{name}/test#
Start a test session. Mints a short-lived session so you can talk to the agent in the browser without a phone call. Returns a room + join token. A test session is a call: it spends one of your plan's simultaneous-call slots for as long as it runs.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
402 |
Too many simultaneous calls for your plan (concurrent_calls_cap): the message names how many your plan allows at once and how many are in progress, counted across your whole account. Nothing was placed or billed; retry when a call ends. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl -X POST "https://api.voiceland.ai/v1/agents/{name}/test" \
-H "Authorization: Bearer VL_API_KEY"
Beispielantwort
{
"expires_at": "2026-07-15T12:00:00Z",
"room_name": "room_…",
"session_id": "sess_…"
}
GET /v1/agents/{name}/versions#
List agent versions. Every save snapshots the agent. Returns the version history.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl "https://api.voiceland.ai/v1/agents/{name}/versions" \
-H "Authorization: Bearer VL_API_KEY"
Beispielantwort
{
"items": [
{
"created_at": "2026-07-14T09:00:00Z",
"version_id": "v_…"
}
]
}
GET /v1/agents/{name}/versions/{version_id}#
Get an agent version. Returns a specific snapshot including its full spec.
Parameter
| Name |
Ort |
Typ |
Erforderlich |
Beschreibung |
name |
path |
string |
Ja |
Agent name. |
version_id |
path |
string |
Ja |
Version id. |
Antworten
| Code |
Beschreibung |
200 |
Success. |
401 |
Missing or invalid API key. |
4XX |
Request error (validation, not-found, etc.). |
5XX |
Server or upstream error. |
Beispielanfrage
curl "https://api.voiceland.ai/v1/agents/{name}/versions/{version_id}" \
-H "Authorization: Bearer VL_API_KEY"
Beispielantwort
{
"spec": {
"name": "front-desk"
},
"version_id": "v_…"
}
Schemata#
Agent#
A voice agent. POST this shape to create or update an agent; server-managed knobs are filled from your account defaults.
| Feld |
Typ |
Erforderlich |
Beschreibung |
calendar_id |
string |
|
|
call_webhook |
CallWebhook |
|
Optional post-call POST: transcript + summary + extracted custom fields + audio URLs, delivered with retry + HMAC signing. |
cdr_export |
CDRExport |
|
Optional lean metadata-only post-call POST for billing / archive systems (no transcript or summary). |
data_sources |
array of string |
|
Ids of data sources (ds-…) whose latest snapshot is placed in the agent's context at call start. Up to 8. |
event_webhooks |
EventWebhooks |
|
Optional real-time per-event delivery (answered / caller utterance / agent utterance / hangup / transfer). |
identity |
Identity |
Ja |
The agent's system prompt and first spoken line, your primary content surface. |
intents |
IntentTaxonomy |
|
|
knowledge_collections |
array of string |
|
|
language |
Language |
Ja |
Primary BCP-47 language (drives speech recognition + the default summary language) plus optional secondary fallbacks. |
name |
string |
Ja |
Logical identifier: lowercase, alphanumeric plus dash/underscore, ≤64 chars. |
notifications |
Notifications |
|
Optional per-call email recap to a list of recipients, independent of call_webhook. |
outbound |
boolean |
|
When true the agent is outbound-only (can be dialed via /calls/dial); when false (default) it is inbound-only and answers calls routed to its numbers. |
pickup_delay_s |
integer |
|
Seconds to wait after answering before the agent speaks its first line (0 to 15). Use 1 to 2 for a more human pace. |
re_engagement_prompt |
string |
|
Line spoken when the caller has been silent for the re-engagement interval. |
record_call |
boolean |
Ja |
When true, per-call stereo + per-channel recordings are produced and their URLs exposed. |
search_widget |
SearchWidget |
|
Optional public knowledge-search box rendered by /search-widget.js on your site. Independent of web_widget: enabling the chat bubble never enables this, and this serves anonymous readers rather than a conversation. |
timezone |
string |
|
IANA timezone for any wall-clock formatting in prompts/tool args. Empty falls back to the account default. |
voice |
Voice |
Ja |
The synthesized voice: provider + voice id (+ optional model). |
web_widget |
WebWidget |
|
Optional browser-embed voice/chat bubble rendered by /embed.js on your site. |
AgentGenerateRequest#
| Feld |
Typ |
Erforderlich |
Beschreibung |
description |
string |
Ja |
Free-text description of what the agent should do. |
language |
string |
|
|
name |
string |
Ja |
Name for the generated agent. |
pickup_delay_s |
integer |
|
|
record_call |
boolean |
|
|
timezone |
string |
|
|
voice_id |
string |
|
|
CDRExport#
| Feld |
Typ |
Erforderlich |
Beschreibung |
headers |
map of string |
|
|
method |
string |
|
|
url |
string |
Ja |
HTTPS endpoint the lean metadata-only CDR payload is delivered to. |
CallWebhook#
| Feld |
Typ |
Erforderlich |
Beschreibung |
custom_fields |
array of CustomField |
|
Fields the model extracts from the transcript and includes in the payload. |
headers |
map of string |
|
Extra request headers sent verbatim (e.g. an Authorization bearer the receiver expects). |
method |
string |
|
HTTP method, POST (default), PUT, or PATCH. |
summary_languages |
array of string |
|
One post-call summary per BCP-47 code; empty uses the agent's primary language. |
url |
string |
Ja |
HTTPS endpoint the post-call payload is delivered to. |
CustomField#
| Feld |
Typ |
Erforderlich |
Beschreibung |
name |
string |
Ja |
Key under which the extracted value is delivered. |
prompt |
string |
Ja |
One-line instruction telling the model what to extract; return null when undeterminable. |
type |
string |
|
Informational type hint (e.g. string). |
EmailNotification#
| Feld |
Typ |
Erforderlich |
Beschreibung |
custom_fields |
array of CustomField |
|
|
summary_languages |
array of string |
|
|
to |
array of string |
Ja |
|
Error#
Error envelope returned for non-2xx responses.
| Feld |
Typ |
Erforderlich |
Beschreibung |
error |
object |
Ja |
|
EventWebhook#
| Feld |
Typ |
Erforderlich |
Beschreibung |
enabled |
boolean |
Ja |
|
headers |
map of string |
|
|
method |
string |
|
|
url |
string |
Ja |
|
EventWebhooks#
Identity#
The agent's system prompt and opening line.
| Feld |
Typ |
Erforderlich |
Beschreibung |
initial_message |
string |
Ja |
The exact first line the agent speaks on pickup (spoken verbatim, not model-generated). |
system_prompt |
string |
Ja |
The master instruction set the model follows for the whole call. |
Intent#
| Feld |
Typ |
Erforderlich |
Beschreibung |
action |
string |
Ja |
|
description |
string |
|
|
examples |
array of string |
|
|
name |
string |
Ja |
|
target |
string |
|
|
IntentTaxonomy#
| Feld |
Typ |
Erforderlich |
Beschreibung |
enabled |
boolean |
Ja |
|
items |
array of Intent |
|
|
threshold |
number |
|
|
unknown_action |
string |
|
|
Language#
| Feld |
Typ |
Erforderlich |
Beschreibung |
primary |
string |
Ja |
Primary language as a BCP-47 tag, e.g. el-GR or en-US. |
secondary |
array of string |
|
Additional BCP-47 languages the agent may switch to. |
Notifications#
RollbackRequest#
| Feld |
Typ |
Erforderlich |
Beschreibung |
version_id |
string |
Ja |
The version to restore. |
Public knowledge-search box for your own site. It serves anonymous visitors, so it is switched on separately from the chat bubble and must name both the sites that may embed it and the collections it may search.
| Feld |
Typ |
Erforderlich |
Beschreibung |
allowed_origins |
array of string |
|
Exact-match list of page origins that may query this widget ("https://example.com", scheme, host, optional port; no path, no wildcard). REQUIRED when enabled. Exact match only, so serving both the apex and www means listing both. |
collections |
array of string |
|
Knowledge-base collection ids this widget may search. REQUIRED when enabled, unlike elsewhere in the API an empty list does not mean "everything published", because for an anonymous reader that default would publish your whole corpus. |
enabled |
boolean |
Ja |
Master switch. While false the public endpoints report the widget as unavailable and answer nothing. |
lang |
string |
|
Default BCP-47 language of the page the widget sits on, used as the retrieval preference when a query does not state one. Empty means no preference. |
no_results_text |
string |
|
Line shown when a search matches nothing. |
placeholder |
string |
|
Placeholder text inside the input. |
primary_color |
string |
|
Accent colour as a hex literal (e.g. #3B82F6). |
result_limit |
integer |
|
Results per query. 0 uses the service default; the service caps the ceiling regardless. |
title |
string |
|
Heading rendered above the search box. |
TransferWebhook#
| Feld |
Typ |
Erforderlich |
Beschreibung |
custom_fields |
array of CustomField |
|
|
enabled |
boolean |
Ja |
|
headers |
map of string |
|
|
method |
string |
|
|
url |
string |
Ja |
|
Voice#
| Feld |
Typ |
Erforderlich |
Beschreibung |
model |
string |
|
Optional voice model/engine override; empty uses the provider default. |
provider |
string |
Ja |
Voice provider identifier. |
voice_id |
string |
Ja |
The voice to use. |
| Feld |
Typ |
Erforderlich |
Beschreibung |
ai_responses_enabled |
boolean |
|
|
allowed_origins |
array of string |
|
|
auto_open |
boolean |
|
|
automation_webhook_url |
string |
|
|
avatar_url |
string |
|
|
busy_text |
string |
|
|
button_main_text |
string |
|
|
button_size |
string |
|
|
button_style |
string |
|
|
button_sub_text |
string |
|
|
chat_enabled |
boolean |
Ja |
|
chat_placeholder |
string |
|
|
chat_tab_label |
string |
|
|
default_mode |
string |
|
|
enabled |
boolean |
Ja |
|
end_confirm_no_text |
string |
|
|
end_confirm_text |
string |
|
|
end_confirm_yes_text |
string |
|
|
header_subtitle |
string |
|
|
header_title |
string |
|
|
layout |
string |
|
|
modal_description |
string |
|
|
modal_title |
string |
|
|
position |
string |
|
|
pre_form |
WebWidgetPreForm |
|
|
primary_color |
string |
|
|
start_button_text |
string |
|
|
takeover |
WebWidgetTakeover |
|
|
voice_connecting_text |
string |
|
|
voice_disconnect_text |
string |
|
|
voice_enabled |
boolean |
Ja |
|
voice_error_text |
string |
|
|
voice_tab_label |
string |
|
|
| Feld |
Typ |
Erforderlich |
Beschreibung |
key |
string |
Ja |
|
label |
string |
Ja |
|
required |
boolean |
|
|
type |
string |
|
|
| Feld |
Typ |
Erforderlich |
Beschreibung |
description |
string |
|
|
fields |
array of WebWidgetFormField |
|
|
submit_text |
string |
|
|
title |
string |
|
|
| Feld |
Typ |
Erforderlich |
Beschreibung |
closed_message |
string |
|
|
enabled |
boolean |
Ja |
|
no_rep_message |
string |
|
|