Documentation Index Fetch the complete documentation index at: https://docs.zyeta.io/llms.txt
Use this file to discover all available pages before exploring further.
The Agents Service provides endpoints for creating, configuring, and deploying AI agents within your organization. It enables you to define agent behaviors, tools, and deployment configurations.
Agent Lifecycle Overview
Authentication
All endpoints require a valid Bearer token in the Authorization header.
Base URL
Agent Endpoints
Create Agent
Create a new AI agent for your organization.
curl
python
typescript
go
Response
curl -X POST {{baseUrl}}/api/agents?org_id=your-org-id \
-H "Authorization: Bearer YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"name": "Customer Support Agent",
"description": "AI agent for handling customer support inquiries",
"agent_type": "chat",
"model_id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
"tools": [
{
"tool_id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
"config": [
{
"name": "api_key",
"value": "sk-123456789"
}
]
}
],
"params": {
"system_prompt": "You are a helpful customer support agent for our e-commerce platform. Your goal is to assist customers with their inquiries, help them track orders, and resolve issues with their purchases. If you don't know the answer, you can use available tools to look up information.",
"temperature": 0.7,
"max_tokens": 1000
},
"is_public": false
}'
Endpoint: POST /api/agents
Query Parameters:
Parameter Required Description
org_idYes Organization ID
Request Body:
Field Type Required Description
namestring Yes Name of the agent descriptionstring Yes Description of the agentβs purpose agent_typestring Yes Type of agent (e.g., βchatβ, βfunctionβ) model_idstring Yes ID of the LLM model to use toolsarray No List of tools to provide to the agent paramsobject Yes Configuration parameters for the agent is_publicboolean No Whether the agent is publicly available (default: false)
Agent Parameters:
Field Type Description
system_promptstring The system instructions for the agent temperaturenumber Sampling temperature (0-1) max_tokensnumber Maximum tokens in the response
List Agents
Retrieve all agents for an organization.
Request
Python
JavaScript
Response
curl -X GET {{baseUrl}}/api/agents/list?org_id=your-org-id \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: GET /api/agents/list
Query Parameters:
Parameter Required Description
org_idYes Organization ID
Get Agent Details
Retrieve detailed information about a specific agent.
Request
Python
JavaScript
Response
curl -X GET {{baseUrl}}/api/agents/get?agent_id=9d8e7f6g-5h4i-3j2k-1l0m-9n8o7p6q5r4s \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: GET /api/agents/get
Query Parameters:
Parameter Required Description
agent_idYes ID of the agent to retrieve
Update Agent
Update an existing agentβs configuration.
Request
Python
JavaScript
Response
curl -X PUT {{baseUrl}}/api/agents/update?agent_id=9d8e7f6g-5h4i-3j2k-1l0m-9n8o7p6q5r4s \
-H "Authorization: Bearer YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"name": "Premium Customer Support Agent",
"description": "Enhanced AI agent for handling premium customer support inquiries",
"params": {
"system_prompt": "You are a helpful premium customer support agent...",
"temperature": 0.5
}
}'
Endpoint: PUT /api/agents/update
Query Parameters:
Parameter Required Description
agent_idYes ID of the agent to update
Request Body:
Field Type Required Description
namestring No New name for the agent descriptionstring No New description model_idstring No New model ID toolsarray No Updated list of tools paramsobject No Updated configuration parameters is_publicboolean No Update public availability
Delete Agent
Delete an agent from the organization.
Request
Python
JavaScript
Response
curl -X DELETE {{baseUrl}}/api/agents/delete?agent_id=9d8e7f6g-5h4i-3j2k-1l0m-9n8o7p6q5r4s \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: DELETE /api/agents/delete
Query Parameters:
Parameter Required Description
agent_idYes ID of the agent to delete
Agent Deployment Endpoints
Create Deployment
Create a new deployment for an existing agent.
Request
Python
JavaScript
Response
curl -X POST {{baseUrl}}/api/agent_deploy/create?org_id=your-org-id \
-H "Authorization: Bearer YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"agent_id": "9d8e7f6g-5h4i-3j2k-1l0m-9n8o7p6q5r4s",
"name": "Web Widget Deployment",
"description": "Customer support agent for the web widget",
"deployment_type": "widget",
"config": {
"theme": {
"primary_color": "#4f46e5",
"font_family": "Inter, sans-serif"
},
"branding": {
"logo_url": "https://example.com/logo.png",
"company_name": "Example Corp"
},
"widget_placement": "bottom-right",
"welcome_message": "π Hi there! How can I help you today?"
}
}'
Endpoint: POST /api/agent_deploy/create
Query Parameters:
Parameter Required Description
org_idYes Organization ID
Request Body:
Field Type Required Description
agent_idstring Yes ID of the agent to deploy namestring Yes Name of the deployment descriptionstring Yes Description of the deployment deployment_typestring Yes Type of deployment (e.g., βwidgetβ, βapiβ, βiframeβ) configobject Yes Configuration parameters for the deployment
List Deployments
Retrieve all deployments for an organization.
Request
Python
JavaScript
Response
curl -X GET {{baseUrl}}/api/agent_deploy/list?org_id=your-org-id \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: GET /api/agent_deploy/list
Query Parameters:
Parameter Required Description
org_idYes Organization ID
Get Deployment Details
Retrieve detailed information about a specific deployment.
Request
Python
JavaScript
Response
curl -X GET {{baseUrl}}/api/agent_deploy/get?deployment_id=5a4b3c2d-1e0f-9g8h-7i6j-5k4l3m2n1o0p \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: GET /api/agent_deploy/get
Query Parameters:
Parameter Required Description
deployment_idYes ID of the deployment to retrieve
Update Deployment
Update an existing deployment configuration.
Request
Python
JavaScript
Response
curl -X PUT {{baseUrl}}/api/agent_deploy/update?deployment_id=5a4b3c2d-1e0f-9g8h-7i6j-5k4l3m2n1o0p \
-H "Authorization: Bearer YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"name": "Web Widget Deployment v2",
"config": {
"theme": {
"primary_color": "#3b82f6"
},
"welcome_message": "π Hello! How may I assist you today?"
}
}'
Endpoint: PUT /api/agent_deploy/update
Query Parameters:
Parameter Required Description
deployment_idYes ID of the deployment to update
Request Body:
Field Type Required Description
namestring No New name for the deployment descriptionstring No New description configobject No Updated configuration parameters
Delete Deployment
Delete a deployment.
Request
Python
JavaScript
Response
curl -X DELETE {{baseUrl}}/api/agent_deploy/delete?deployment_id=5a4b3c2d-1e0f-9g8h-7i6j-5k4l3m2n1o0p \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: DELETE /api/agent_deploy/delete
Query Parameters:
Parameter Required Description
deployment_idYes ID of the deployment to delete
Regenerate API Key
Regenerate the API key for a deployment.
Request
Python
JavaScript
Response
curl -X POST {{baseUrl}}/api/agent_deploy/regenerate_key?deployment_id=5a4b3c2d-1e0f-9g8h-7i6j-5k4l3m2n1o0p \
-H "Authorization: Bearer YOUR_TOKEN"
Endpoint: POST /api/agent_deploy/regenerate_key
Query Parameters:
Parameter Required Description
deployment_idYes ID of the deployment for which to regenerate the API key
Error Responses
Status Code Description
400 Bad Request - Invalid input or validation error 401 Unauthorized - Invalid or missing token 403 Forbidden - Insufficient permissions 404 Not Found - Resource doesnβt exist 409 Conflict - Resource already exists or conflict with existing resource 500 Internal Server Error - Server-side error
Implementation Notes
Agents are configured with specific LLM models, tools, and parameters
Agents can be deployed in multiple ways (widget, API, SDK)
Each deployment has its own API key for authentication
Deployments can be customized with different themes and configurations
Usage statistics are available for each deployment
Organization-based access control ensures separation between tenants