> ## Documentation Index > Fetch the complete documentation index at: https://docs.nugen.in/llms.txt > Use this file to discover all available pages before exploring further. # Create Agent Endpoint > Create a new AI agent with custom configuration. This endpoint creates a new agent that can be customized with specific instructions, demonstrations, and model parameters. **Request Body:** - `agent_name`: Name of the agent (will be normalized to URL-friendly format) - `agent_description`: Description of the agent's purpose - `model`: Model ID (base model or organization's aligned model) - `max_tokens`: Maximum tokens for agent responses - `temperature`: Sampling temperature for response generation (0.0-2.0, default: 0.7) - `instructions` (optional): System prompt/instructions - `demonstrations` (optional): List of example interactions for few-shot learning **Returns:** - `data`: Agent data object with the following fields: - `id`: Unique agent identifier - `name`: Normalized agent name (lowercase with underscores) - `description`: Agent description - `base_model`: Model ID used by the agent - `base_model_name`: Human-readable model name - `created_date`: ISO 8601 timestamp of creation - `status`: Agent status (e.g., "ACTIVE") - `creator`: Username/email of the agent creator - `endpoint`: Full API endpoint URL for running the agent - `usage_count`: Initial count (0) - `last_used`: Last usage timestamp (null for new agents) - `performance_metrics`: Performance statistics - `system_prompt`: Agent instructions - `temperature`: Temperature for model responses (0.0-2.0) - `conversation_saving_enabled`: Whether conversations are saved **Errors:** - `400`: Requested model is not available to the user's organization - `401`: User is not found or unauthorized **Example Request:** ```json POST /api/v3/agents/create Headers: {"Authorization": "Bearer "} { "agent_name": "Customer Support Bot", "agent_description": "Handles customer inquiries with empathy", "model": "nugen-flash-instruct", "max_tokens": 2000, "temperature": 0.7, "instructions": "You are a helpful customer support agent. Be empathetic and solution-oriented.", "demonstrations": [ { "input": "How do I reset my password?", "output": "I'd be happy to help you reset your password..." } ] } ``` **Example Response:** ```json { "data": { "id": "agent-abc123", "name": "customer_support_bot", "description": "Handles customer inquiries with empathy", "base_model": "nugen-flash-instruct", "base_model_name": "Nugen Flash Instruct", "created_date": "2024-01-15T10:30:00Z", "status": "ACTIVE", "creator": "user@example.com", "endpoint": "https://api.nugen.in/api/v3/agents/run-agents/customer_support_bot/run", "usage_count": 0, "last_used": null, "performance_metrics": { "total_requests": 0, "average_response_time": 0, "success_rate": 0 }, "system_prompt": "You are a helpful customer support agent. Be empathetic and solution-oriented.", "temperature": 0.7, "conversation_saving_enabled": false } } ``` ## OpenAPI ````yaml https://api.nugen.in/openapi-public.json post /api/v3/agents/create openapi: 3.1.0 info: title: Nugen Intelligence API description: 'Nugen Intelligence : Powering Specialized Intelligence At Scale' version: 25.4.20 servers: [] security: [] paths: /api/v3/agents/create: post: tags: - Agents summary: Create Agent Endpoint description: >- Create a new AI agent with custom configuration. This endpoint creates a new agent that can be customized with specific instructions, demonstrations, and model parameters. **Request Body:** - `agent_name`: Name of the agent (will be normalized to URL-friendly format) - `agent_description`: Description of the agent's purpose - `model`: Model ID (base model or organization's aligned model) - `max_tokens`: Maximum tokens for agent responses - `temperature`: Sampling temperature for response generation (0.0-2.0, default: 0.7) - `instructions` (optional): System prompt/instructions - `demonstrations` (optional): List of example interactions for few-shot learning **Returns:** - `data`: Agent data object with the following fields: - `id`: Unique agent identifier - `name`: Normalized agent name (lowercase with underscores) - `description`: Agent description - `base_model`: Model ID used by the agent - `base_model_name`: Human-readable model name - `created_date`: ISO 8601 timestamp of creation - `status`: Agent status (e.g., "ACTIVE") - `creator`: Username/email of the agent creator - `endpoint`: Full API endpoint URL for running the agent - `usage_count`: Initial count (0) - `last_used`: Last usage timestamp (null for new agents) - `performance_metrics`: Performance statistics - `system_prompt`: Agent instructions - `temperature`: Temperature for model responses (0.0-2.0) - `conversation_saving_enabled`: Whether conversations are saved **Errors:** - `400`: Requested model is not available to the user's organization - `401`: User is not found or unauthorized **Example Request:** ```json POST /api/v3/agents/create Headers: {"Authorization": "Bearer "} { "agent_name": "Customer Support Bot", "agent_description": "Handles customer inquiries with empathy", "model": "nugen-flash-instruct", "max_tokens": 2000, "temperature": 0.7, "instructions": "You are a helpful customer support agent. Be empathetic and solution-oriented.", "demonstrations": [ { "input": "How do I reset my password?", "output": "I'd be happy to help you reset your password..." } ] } ``` **Example Response:** ```json { "data": { "id": "agent-abc123", "name": "customer_support_bot", "description": "Handles customer inquiries with empathy", "base_model": "nugen-flash-instruct", "base_model_name": "Nugen Flash Instruct", "created_date": "2024-01-15T10:30:00Z", "status": "ACTIVE", "creator": "user@example.com", "endpoint": "https://api.nugen.in/api/v3/agents/run-agents/customer_support_bot/run", "usage_count": 0, "last_used": null, "performance_metrics": { "total_requests": 0, "average_response_time": 0, "success_rate": 0 }, "system_prompt": "You are a helpful customer support agent. Be empathetic and solution-oriented.", "temperature": 0.7, "conversation_saving_enabled": false } } ``` operationId: create_agent requestBody: content: application/json: schema: $ref: '#/components/schemas/CreateAgentRequest' required: true responses: '200': description: >- Returns the details of the newly created agent including its unique identifier, configuration, usage statistics, and endpoint URL for execution content: application/json: schema: $ref: '#/components/schemas/AgentCreateResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: CreateAgentRequest: properties: agent_name: type: string title: Agent Name description: Name of the agent example: Customer Support Agent agent_description: type: string title: Agent Description description: Description of the agent example: An agent designed to handle customer support queries model: type: string title: Model description: Base model identifier example: nugen-flash-instruct max_tokens: anyOf: - type: integer - type: 'null' title: Max Tokens description: Maximum number of tokens to generate in the response default: 500 example: 500 temperature: type: number maximum: 2 minimum: 0 title: Temperature description: Temperature for model responses (0.0-2.0) default: 0.7 example: 0.7 instructions: anyOf: - type: string - type: 'null' title: Instructions description: >- System prompt for the agent (optional, defaults to 'You are a helpful AI assistant.') example: You are a helpful customer support assistant. demonstrations: anyOf: - items: additionalProperties: type: string type: object type: array - type: 'null' title: Demonstrations description: Training demonstrations type: object required: - agent_name - agent_description - model title: CreateAgentRequest AgentCreateResponse: properties: data: $ref: '#/components/schemas/AgentData' type: object required: - data title: AgentCreateResponse description: Response model for agent creation endpoint. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError AgentData: properties: id: type: string title: Id description: Unique agent identifier example: agent-12345 name: type: string title: Name description: Agent name example: Customer Support Agent description: type: string title: Description description: Agent description example: An agent designed to handle customer support queries base_model: type: string title: Base Model description: Base model identifier example: nugen-flash-instruct base_model_name: type: string title: Base Model Name description: Base model display name example: Nugen Flash Instruct created_date: type: string title: Created Date description: Agent creation date example: '2024-01-15T10:30:00Z' status: $ref: '#/components/schemas/AgentStatus' description: Agent status example: ACTIVE creator: type: string title: Creator description: Agent creator username example: youremail@domain.com endpoint: type: string title: Endpoint description: Agent API endpoint example: https://api.nugen.in/api/v3/inference/completions/agent-12345 usage_count: type: integer title: Usage Count description: Number of times agent has been used example: 150 last_used: anyOf: - type: string - type: 'null' title: Last Used description: Last usage timestamp example: '2024-01-20T14:30:00Z' performance_metrics: $ref: '#/components/schemas/AgentPerformanceMetrics' description: Agent performance metrics system_prompt: type: string title: System Prompt description: System prompt for the agent example: You are a helpful customer support assistant. temperature: type: number title: Temperature description: Temperature for model responses (0.0-2.0) example: 0.7 conversation_saving_enabled: type: boolean title: Conversation Saving Enabled description: Whether conversation saving is enabled for this agent default: false example: true type: object required: - id - name - description - base_model - base_model_name - created_date - status - creator - endpoint - usage_count - performance_metrics - system_prompt - temperature title: AgentInfo ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type type: object required: - loc - msg - type title: ValidationError AgentStatus: type: string enum: - ACTIVE - INACTIVE title: AgentStatus AgentPerformanceMetrics: properties: average_uncertainty: type: number title: Average Uncertainty description: Average uncertainty across all agent responses example: 0.15 human_interventions: type: integer title: Human Interventions description: Number of times human intervention was required example: 5 average_response_time: type: number title: Average Response Time description: Average response time in seconds example: 1.2 user_satisfaction_score: type: number title: User Satisfaction Score description: Average user satisfaction score from 0.0 to 1.0 example: 0.85 type: object required: - average_uncertainty - human_interventions - average_response_time - user_satisfaction_score title: AgentPerformanceMetrics securitySchemes: HTTPBearer: type: http scheme: bearer ````