How to Safely Interpret Added VS Code Agents in Copilot Usage Metrics

PowerShellカテゴリを表すパンダのイラスト PowerShell

This article is a technical explanation and implementation example created using AI. Although the code and procedures provided are based on primary sources, the author has not verified their operation on actual devices. Behavior may vary depending on the environment and version. , We outline the requirements and mechanisms for safely understanding changes related to VS Code Agents in GitHub Copilot usage metrics and accurately interpreting usage conditions across organizations and enterprises. The purpose is to check the structure of metrics provided in official changelogs and the specifications for handling them in APIs and aggregation reports in advance before performing verification in an actual device environment.


1. Purpose

The purpose of this article is to correctly understand the newly integrated VS Code Agent usage statistics data items based on official GitHub Copilot usage metrics announcements. We organize the metric structures used by organization and enterprise administrators and billing managers to measure adoption and engagement, and clarify precautions for use and prerequisites for access privileges.

2. Prerequisites and Precautions

  • [Planned to be verified in a Windows environment] or [Before actual device verification]: This article does not show the results of directly retrieving the output of the APIs and aggregation items explained here using PowerShell or CLI in a real environment.

  • Limitation of target features: The metrics here target activity within the dedicated VS Code Agents window. They are managed separately from Agent Mode within the editor window and general-purpose rollups.

  • Privilege management: Accessible only to enterprise owners, billing managers, organization owners, or users assigned a custom role with the "View Copilot Metrics" permission. In addition, the Copilot usage metrics policy must be enabled.

  • Behavior when data is absent: Backward compatibility is maintained, and if data for the VS Code Agents window does not exist, optional fields will be absent or null will be returned.

3. Structure of Added Metrics

According to primary sources, enterprise and organization summary reports (1-day and 28-day periods) now include new metrics concerning the dedicated VS Code Agents window. Here, we examine the added key elements individually.

flowchart TD
    A[Copilot Usage Metrics] --> B[Aggregate Reports: 1-day / 28-day]
    A --> C[User-Level Reports: Enterprise / Org]

    B --> B1[daily_active_vscode_agent_users: 1日あたりのユニークユーザー数]
    B --> B2[totals_by_vscode_agent: session_count と total_user_messages の集計]

    C --> C1[used_vscode_agent: 利用有無を示すオプショナルな指標]
    C --> C2[totals_by_vscode_agent: ユーザーごとのセッション数とメッセージ数]

daily_active_vscode_agent_users

An optional count item in aggregation reports indicating the number of unique users active daily in the dedicated VS Code Agents window. It is used to gauge the scale of daily usage across the organization.

totals_by_vscode_agent (Aggregation Reports)

Optional combined values for session_count (number of sessions) and total_user_messages (total number of user messages) regarding VS Code Agents. This allows measuring the depth of engagement through message interaction and session frequency, rather than just headcount.

used_vscode_agent

An item included in enterprise and organization user-level reports (1-day and 28-day periods). It functions as an optional indicator showing whether individual users have utilized the VS Code Agents window.

totals_by_vscode_agent (User Level)

The values for session_count and total_user_messages per individual user in user-level reports. This is useful for closely understanding usage variations among teams and the degree of utilization in specific teams.

4. Key Points for API and Report Utilization

When utilizing the official Copilot usage metrics API, the following points should be noted:

  • Selection of reporting periods: Since both 1-day and 28-day periods are provided, you can track both short-term trend changes and medium-to-long-term adoption.

  • Understanding data separation: Metrics for the VS Code Agents window are clearly distinguished from Agent Mode within the editor window and traditional general-purpose aggregations. Therefore, care must be taken not to confuse them with overall aggregated values.

  • Policy and access settings: Before calling the API, make sure that the Copilot usage metrics policy is properly enabled on GitHub and that the executing user is assigned the necessary privileges (such as View Copilot Metrics).

5. Limitations

  • The content covered in this article is a specification explanation based on the official changelog as of September 11, 2026, and field names or behavior may change due to future API updates or feature expansions.

  • Because specific JSON outputs of API responses in an actual device environment and measured values recorded via scripts using PowerShell or similar tools have not been performed, please refer directly to the official Copilot usage metrics API documentation when deploying.

6. Conclusion

  • Points to check before execution:

    • Whether the Copilot usage metrics policy is enabled for the organization or enterprise.

    • Whether the accessing user is granted "View Copilot Metrics" permissions or appropriate owner privileges.

    • Whether the data being retrieved targets the dedicated VS Code Agents window rather than the regular Editor Agent Mode.

  • Constraints:

    • The contents of this article are based on research without actual device verification, and behavior may vary depending on differences in environments and API versions.

    • The absence of optional fields or the behavior of null values occurs when data is absent based on backward compatibility, which must be accounted for in the application's parsing logic.

Reference Information

Document information

Article title
How to Safely Interpret Added VS Code Agents in Copilot Usage Metrics
Published
Updated
Source
https://papanda925.com/?p=16048&lang=en

License: Text and original figures for which this site holds the relevant rights are available under CC BY 4.0 , unless otherwise noted. This article may include content created or edited with generative AI. If code has a separate license notice or a linked GitHub repository license, that license takes precedence for the code. Quotations, third-party materials, images, and trademarks are excluded from this license. Usage policy

Copied title and URL