Agent Console Guide

This page provides a guide to the Agent entity in the Namirasoft Inference Console. It defines the concepts and configuration fields used when creating and managing Agents. Use this guide to understand the purpose and behavior of each setting available during Agent configuration.

What Is an Agent?

An Agent in Namirasoft Inference is a configuration that defines an AI model and how it is accessed. Each Agent specifies a single AI model, the way that model is accessed, either through Namirasoft or through your own key, and the cost limits applied to its use.

 

The model you configure in an Agent is used by supported Namirasoft applications, such as Namirasoft Expert. When a supported application needs to use an AI model, it uses the configuration defined in your Agent. The content submitted to the model is processed by those applications and the applicable model provider, not by Namirasoft Inference.

 

For information about available models and how access and pricing work, visit the Pricing page.

The Challenge in Managing AI Models

Teams that use AI across more than one application face the challenge of managing models, providers, and cost in a consistent way. Without a central place to define how AI is accessed, each application has to be set up on its own, and usage becomes difficult to oversee.

 

Common challenges include:

 

• Duplicated setup: Each application may require its own model selection and provider configuration, repeating the same work in several places.
• API key management: Using models directly often requires creating and maintaining a separate account and API key with each provider.
• Limited cost control: When usage is spread across applications and providers, keeping spending within expected limits becomes difficult.
• Difficult model changes: Switching to a different model or provider can require reworking each application that uses it.
• Model selection overhead: Choosing the right model for each use case means comparing options across several providers.

How Namirasoft Inference Solves the Problem

Namirasoft Inference addresses these challenges through Agents. An Agent is an individually configured definition of an AI model and how it is accessed, together with the cost limits applied to its use.

 

Instead of configuring AI separately in every application, you define an Agent once in Namirasoft Inference, and supported Namirasoft applications use that configuration when they need a model. You can use Namirasoft provided access without managing an API key, or connect your own provider credential, and you can change the model without reworking the applications that rely on it.

 

Cost limits set on an Agent apply automatically, and usage totals are tracked for each Agent, giving you visibility and control over how AI is used across your applications.

Overview of Agent Fields and Options

Below is a detailed explanation of the fields available when creating or managing an Agent. Understanding these fields helps ensure your Agent is configured correctly for your requirements.

• ID (String): This is a unique identifier automatically assigned to the Agent when it is created. The system uses it to track, reference, and manage this specific Agent. This value is auto-generated and cannot be modified.

• User ID (Namirasoft Account’s ID): This is the unique identifier of the Namirasoft Account user who owns this Agent. It is used internally for permission control, audit logging, and access management.

• Workspace ID (Namirasoft Workspace’s ID): This is the identifier of the workspace this Agent belongs to, as defined in Namirasoft Workspace. A workspace is a shared organizational space where teams group their agents, projects, and members.

• Name (String): This is a label used to identify this Agent in the console. A good name clearly describes the Agent’s purpose or the model it is configured to use. For example: “Production Assistant” or “Fast Everyday Model”.

• Model Company (Enum): This is the provider of the AI model this Agent uses. Supported companies are OpenAI, Anthropic, DeepSeek, and Namirasoft (for Namirasoft Squirrel). Selecting the company determines which models are available to choose from.

• Model Name (Enum): This is the specific model, from the selected Model Company, that this Agent uses. The available models depend on the chosen company and are displayed in the console.

• Model API Key Credential (Credential ID, optional): When set, the Agent accesses the model using your own provider credential, an arrangement also described as bring your own key. When left empty, the Agent uses Namirasoft-provided access and no separate API key is required. Namirasoft Squirrel is available exclusively through Namirasoft.

• Maximum Cost Per Run (USD, optional): Maximum Cost Per Run limits how much this Agent can spend on a single run. Once the limit is reached, the run stops.

• Maximum Cost Per Chat (USD, optional): Maximum Cost Per Chat limits how much this Agent can spend in a single chat session. Once the limit is reached, the session stops accepting new requests.

• Maximum Cost Per Day (USD, optional): Maximum Cost Per Day limits how much this Agent can spend in a single calendar day. Resets automatically at the start of each day.

• Maximum Cost Per Week (USD, optional): Maximum Cost Per Week limits how much this Agent can spend in a single calendar week. Resets automatically at the start of each week.

• Maximum Cost Per Month (USD, optional): Maximum Cost Per Month limits how much this Agent can spend in a single calendar month. Resets automatically at the start of each month.

• Created At (DateTime): This is the date and time when this Agent was created. This value is automatically generated and cannot be modified.

• Updated At (DateTime): This is the date and time when this Agent configuration was last modified. This value is updated automatically whenever any field is changed.