ORGN Usage & Metrics Overview

The Usage dashboard provides visibility into how ORGN is being used across your team and projects. It aggregates agent activity, token consumption, and cost metrics into a single view so you can monitor adoption, understand resource utilization, and track operational impact over time.

This dashboard is especially useful for engineering leads, platform owners, and finance stakeholders who need transparency into how AI-assisted development is being consumed.

All data shown reflects the currently selected team and time range.

Summary Metrics

At the top of the page, four summary cards provide a high-level snapshot of overall activity.

Agent Runs

The Agent Runs metric represents the total number of times an agent executed structured work during the selected time period.

An agent run is recorded when the system performs an execution-oriented action such as:

• Writing or modifying code
• Running terminal commands
• Performing structured repository analysis
• Executing a workflow like QA, SecOps, or Scan & Review

This metric reflects how actively AI capabilities are being used. A higher number of agent runs typically indicates deeper AI integration into development workflows.

For example, a spike in agent runs may correspond to a major feature build, a refactor, or heavy documentation generation.

Sessions

The Sessions metric shows how many interactive sessions were created during the selected period.

A session represents a working conversation between a user and an agent. It may include:

• Instructions and clarifications
• File reads and edits
• Command execution
• Diff generation
• Review workflows

This number helps you understand how frequently team members are engaging with ORGN’s workspace. A small number of sessions with very high token usage may indicate long-running or complex interactions.

Total Tokens

Total Tokens reflects the cumulative number of model tokens consumed across all sessions and agent runs.

This includes:

• Input tokens (prompts and attached context)
• Output tokens (agent responses and reasoning)
• Internal reasoning tokens used during execution

Token usage is the primary unit of AI resource consumption. It correlates directly with model workload intensity.

High token consumption may be associated with:

• Large repository analysis
• Complex architectural reasoning
• Long documentation generation
• Large diff production

Total Cost

The Total Cost metric displays the monetary cost associated with token usage during the selected period.

This value aggregates:

• Model inference costs
• Provider-specific pricing differences
• Confidential compute pricing (if applicable)

The cost metric allows teams to:

• Monitor AI spending
• Identify cost spikes
• Align usage with budgets
• Evaluate ROI of AI-assisted workflows

Token Consumption by Project

The Token Consumption by Project chart visualizes how token usage is distributed across projects over time.

Time Filter

In the top-right corner of this section, you can select the time window (for example, “Last 7 days”). All dashboard metrics update dynamically based on this selection.

Project Breakdown

Each bar represents the number of tokens consumed by a specific project on a given date.

This chart helps answer questions such as:

• Which projects are consuming the most AI resources?
• Did a particular release or refactor drive increased usage?
• Is usage steady, growing, or declining over time?

For example, if one project shows a sharp spike in token usage on a specific date, it may indicate:

• A large refactor
• A documentation generation workflow
• A significant AI-assisted feature implementation
• A security or QA review pass

This view makes it easier to correlate usage with development activity.

Usage Leaderboard

The Usage Leaderboard ranks team members by activity during the selected time period.

This section provides a user-level breakdown of engagement and contribution.

Each row includes the following fields:

User

Displays the team member’s name and profile identity. This allows you to track usage distribution across contributors.

Accepted Diffs

Accepted Diffs indicates how many AI-generated code changes were reviewed and approved by the user.

This metric reflects:

• Trust in AI output
• Adoption of AI-assisted coding
• Volume of accepted contributions

A high number of accepted diffs suggests that AI-generated changes are frequently making it into the project’s working branch.

Tasks Completed

This shows how many structured tasks were marked complete by the user.

Tasks represent defined objectives such as:

• Implementing a feature
• Fixing a bug
• Generating documentation
• Completing a review

This metric reflects workflow throughput rather than raw token consumption.

Tokens

Displays the total tokens consumed by that user’s sessions and agent runs.

This helps identify:

• Heavy AI users
• Power users of agent workflows
• Contributors working on large or complex tasks

Cost

Shows the cost associated with the user’s token consumption.

This can help teams:

• Monitor per-user AI cost distribution
• Identify usage concentration
• Evaluate training or adoption patterns

Agent Lines of Code

This metric reflects the total number of lines of code generated or modified by agents and accepted by the user.

It provides insight into:

• The scale of AI-assisted development
• Codebase expansion or restructuring
• Refactor intensity

For example, a high number of agent-generated lines combined with many accepted diffs suggests substantial AI-driven implementation work.

Interpreting the Dashboard Holistically

While each metric provides standalone insight, the dashboard becomes most valuable when metrics are interpreted together.

Examples:

• High agent runs + high accepted diffs → Active and trusted AI-assisted development
• High tokens + low tasks completed → Possibly exploratory work or complex debugging
• High cost concentrated in one project → Consider reviewing model selection or workflow efficiency
• High lines of code + low accepted diffs → AI output may require refinement or stronger review controls

By examining patterns rather than individual numbers, teams can better understand AI adoption and efficiency.

What This Dashboard Covers

The Usage dashboard reflects:

• Activity within the selected team
• Activity within the selected time window
• AI-driven execution and token consumption
• Model usage costs

It does not include:

• GitHub activity performed outside ORGN
• Manual commits not tied to sessions
• External development outside the managed workspace

Why the Usage Dashboard Matters

The Usage dashboard provides measurable visibility into AI-assisted development. It ensures that:

• Adoption remains transparent
• Costs are trackable
• Resource consumption is visible
• AI-generated output is accountable

Rather than treating AI usage as opaque, the dashboard makes it observable and auditable across teams and projects.

It transforms AI from a black box into a measurable operational layer of your engineering workflow.