Use the Speak AI MCP server tool create_voice_agent to create a voice agent.
What does it do?
Create a voice agent. Requires the OWNER or ADMIN role. name, personality,
instructions, and voice (provider + voiceId) are required; everything else can be
set now or later with update_voice_agent. Call list_voice_avatars or
list_voices first to get valid ids.
Parameters
create_voice_agent takes 9 parameters, 4 of them required.
| Name | Type | Required | Description |
|---|---|---|---|
name |
string | Yes | Required on create. Trimmed, non-empty. |
personality |
string | Yes | Required on create. Free text describing the agent’s tone. |
instructions |
string | Yes | Required on create. The agent’s system instructions. |
voice |
object | Yes | Required on create. |
llm |
object | No | |
avatar |
object | No | Set to attach a video avatar; avatarUrl/provider are derived server-side from the catalog row. |
conversationMode |
enum: voice_only, video_avatar | No | |
folderId |
string | No | Folder to file this agent’s conversations under. |
enableWebSearch |
boolean | No | Let the agent search the web mid-call, separate from any attached knowledge base. |
Behaviour
The Speak AI MCP server publishes these behaviour hints for create_voice_agent.
- Can change data in your workspace.
- Does not delete or overwrite existing data.
- Repeating an identical call can have an extra effect.
- Stays inside Speak AI. Does not reach other systems.
Related
- Voice agents lists the rest of the Voice agents category.
- Tool reference lists every Speak AI MCP server tool.
- MCP server covers setup, supported clients, and what agents can do once connected.