Skip to content

Tools

Copilot use in OpenCode now shows up in GitHub usage reports

3-line summary

  1. GitHub added Copilot model usage from tools such as OpenCode to reports on October 23.
  2. Reports separate requests, sessions and tokens by tool; requests are not user prompts.
  3. Authorized enterprises and organizations can access daily and 28-day reports with metrics enabled.
PracticalDevelopersOffice workers
Arrows from terminal and editor windows lead to a report with three bars on a white card against a green background
Illustration: AI Ping

GitHub expanded Copilot usage reporting on October 23. Server-side counts now include tools such as OpenCode when they use Copilot as their model provider.

New totals_by_client entries show successful model calls, sessions and input/output tokens. Tool labels follow request integration IDs and are best-effort groupings, not verified identities.

The change covers enterprise, organization and their per-user daily and 28-day reports. The REST reference says reports are generated daily for processed periods. Access requires enterprise owner or billing-manager status, organization ownership, or the relevant metrics-viewing permission, with the Copilot usage metrics policy enabled.

A request_count measures model calls: one prompt can trigger several. Agent-app jobs started on GitHub are a separate measure. Existing IDE metrics also depend on client telemetry and processing time, so comparisons should align reporting periods, scope and data readiness first.

Why it matters

AI Ping sees a stronger basis for deciding which developer tools to support beyond the editor. Teams should agree on what each metric counts before treating traffic as evidence of work outcomes.

Who it's for

Developers
Compare requests and tokens over matching periods, then review actual work results separately.
Office workers
Label prompts, model calls and agent jobs as separate units in internal usage reports.

Sources

  • How-to guideAbout 19 min reading

    Measure an AI pilot and decide when to stop

    Count review and repair as well as generation. Compare fictional task records with a baseline and decide whether to continue, revise or stop.

    AdvancedPlanningDesignersMarketersSalesProduct researchOffice workersDevelopers
  • How-to guideAbout 17 min reading

    Getting good results from AI coding assistants

    Give context, work in small steps, verify with tests — a workflow that makes AI-written code something you can trust.

    Intermediate · practicalDevelopers