v3
OpenAPI 3.1.02026-07-31350397683.8 KBUpdate Agent
Update an existing agent in the system.
Headers
API key required to authenticate requests.
Request body
Stable code-addressable identifier
ID of the agent to be updated
External agent ID for the agent
Name of the agent
System prompt for the agent
Phone number associated with the agent
List of goals for the agent
Enum representing the type of agent
Enum representing the connection type of the agent
Enum representing the mode of agent
List of keyterms for the agent
E.164 campaign numbers authorized to send this agent's follow-up SMS
Name of the folder to put the agent in
WebSocket URL for real-time communication
Username for WebSocket authentication
Password for WebSocket authentication
SIP URI for VoIP calls
Username for SIP authentication
Password for SIP authentication
Custom headers for SIP requests (Do not include X- prefix for the headers, these get added automatically)
Per-call SIP endpoint webhook. When set (SIP connection), each call fetches its destination SIP URI from this URL at dial time.
Customer-dialed sims: Bluejay POSTs a rendered payload to this URL at dispatch so the customer's own flow dials their agent into the receiving digital human.
JSON template for the outbound trigger webhook body; {{tokens}} like {{digital_human.phone_number}} are filled per call.
Enable/disable PII redaction
LiveKit agent name
LiveKit metadata
LiveKit customer tool topic
Webhook URL Bluejay calls per simulated call to fetch a LiveKit room token; alternative to storing the org's LiveKit API key/secret. Empty string clears the field.
Pipecat agent name
Pipecat agent configuration
Pipecat room properties
HTTP webhook URL for text agent communication
Voice provider for the agent
Provider ID for the agent
ElevenLabs Convai agent_id (for ELEVEN_LABS connection type; the xi-api-key lives in the org's integrations row)
Retell agent_id (for RETELL connection type; the retell_key lives in the org's integrations row)
Vapi assistant_id (for VAPI connection type; the vapi_private_key lives in the org's integrations row)
Google CES app id (for GOOGLE_CES_VOICE / GOOGLE_CES_CHAT; the WIF JSON lives in the org's integrations row)
Google CES deployment id (deployment is bound to a specific app version)
Google Dialogflow CX GCP project ID (for GOOGLE_DFCX_VOICE / GOOGLE_DFCX_CHAT; the WIF JSON lives in the org's integrations row)
Google Dialogflow CX location (e.g. us-central1, global)
Google Dialogflow CX agent UUID (from the DFCX agent resource)
Amelia deployment base URL including context root (e.g. https://tenant.amelia.com/Amelia)
Amelia domain code the conversation opens against (from /api/init -> domains[].code)
Which conversation flow in the Amelia domain this agent tracks (required for workflow sync)
Which cognitive agent in the Amelia domain this agent tracks (takes precedence over amelia_flow_name for workflow sync)
Kore.ai platform host (e.g. https://kb.est.k8s.qa.kore.aws.spctrm.net)
Kore.ai bot ID (e.g. st-d536681d-...)
Kore.ai webhook app Client ID (cs-...)
Kore.ai webhook app Client Secret (HS256); stored encrypted
Automatically initiate an outbound call when a test result reaches READY state
Enable/disable simulations for this agent
Provider-specific workflow definition (e.g. full ElevenLabs Get Agent response). Parsing depends on provider.
Source platform the agent's definition (system prompt, workflow) comes from.
Distinct from voice_provider (TTS/STT vendor) and provider_id (handle on that vendor) — this identifies where the agent's behavior (prompt, multi-agent graph) is authored.
Response
Successful Response
Status of the response
Signing key for HTTP webhook (only provided if http_webhook was set)