> ## Documentation Index
> Fetch the complete documentation index at: https://docs.condense.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Agent Usage

# Agent Usage

The Agent Usage page provides visibility into AI token consumption across your Condense environments. It enables administrators to monitor how AI Agents are being used, identify usage trends, and understand token consumption across users, workspaces, and environments.

Every AI request processed by Condense is automatically recorded, allowing organizations to track operational usage and manage AI costs effectively.

## Overview

Agent Usage provides detailed insights into:

* Total AI requests processed
* Input token consumption
* Output token consumption
* Token usage by AI Agent
* Usage by environment
* Usage by workspace
* Usage by user
* Usage over time

This information helps administrators monitor AI adoption and optimize token utilization.

***

## Usage Dashboard

The dashboard provides a consolidated view of AI activity across the organization.

Typical metrics include:

| Metric         | Description                                                   |
| -------------- | ------------------------------------------------------------- |
| Total Requests | Total AI requests processed during the selected period.       |
| Input Tokens   | Total prompt tokens consumed.                                 |
| Output Tokens  | Total response tokens generated.                              |
| Total Tokens   | Combined input and output token consumption.                  |
| Active Users   | Number of users who utilized AI Agents.                       |
| Active Agents  | AI Agents that processed requests during the selected period. |

***

## Usage History

Every AI request generates a usage record.

Each record typically includes:

* Timestamp
* User
* Environment
* Workspace
* AI Agent
* AI Model
* Input Tokens
* Output Tokens
* Total Tokens
* Request Status

Usage history can be filtered and searched for auditing or operational analysis.

***

## Filtering Usage

Usage data can be filtered using one or more criteria.

Available filters include:

* Date Range
* Environment
* Workspace
* User
* AI Agent
* AI Model
* Request Status

Filtering helps isolate specific workloads or investigate abnormal usage patterns.

***

## Usage by Agent

Administrators can view token consumption for each AI Agent.

This helps identify:

* Frequently used agents
* High token-consuming agents
* Low adoption agents
* Overall AI utilization trends

These insights assist with operational planning and cost optimization.

***

## Usage by User

Usage reports can also be grouped by user.

For each user, Condense displays:

* Number of AI requests
* Input tokens consumed
* Output tokens consumed
* Total token usage

This enables administrators to understand adoption across teams and departments.

***

## Usage by Environment

Organizations operating multiple environments can monitor AI usage independently for each environment.

This allows administrators to:

* Compare environments
* Track development versus production usage
* Monitor environment-specific token consumption

***

## Export Usage Reports

Usage reports can be exported for further analysis or auditing.

Supported export formats include:

* CSV
* Excel

Exports contain the currently filtered usage records.

***

## Relationship with Token Wallets

Every successful AI request automatically deducts tokens from the corresponding Input and Output Token Wallets.

The Agent Usage page provides a detailed breakdown of these transactions, while the Token Wallets page displays the remaining balances.

***

## Permissions

Access to Agent Usage is controlled using role-based access control (RBAC).

Depending on assigned permissions, users may be able to:

* View usage dashboards
* View detailed usage history
* Export usage reports
* Monitor organization-wide consumption

Organization-wide usage data is available only to authorized administrators.

***

## Best Practices

* Monitor usage regularly to understand AI adoption.
* Review high token-consuming agents periodically.
* Export reports for billing reconciliation when required.
* Use filters to investigate unusual usage spikes.
* Monitor token consumption alongside wallet balances to avoid service interruptions.

***

## Related Pages

* Billing
* Token Wallets
* Recharge Tokens
* Pricing & Billing