Copilot usage metrics API now counts requests from tools like OpenCode

GitHub says the Copilot usage metrics API now reports model usage from tools outside the IDE that use Copilot as their model provider, including OpenCode, which earlier reports did not capture. Those tools often send no editor telemetry, so their usage was missing from totals_by_ide.

• New field: a totals_by_client array in the enterprise, organization, enterprise-user and organization-user reports, in both 1-day and 28-day versions. In the 28-day reports it appears in each day_totals entry.
• What each entry holds: the client (such as opencode, claude_code, codex, zed, copilot_cli, github_app, cloud_agent, code_review, vscode or jetbrains; unrecognized clients are grouped as other), request_count, session_count, token usage (prompt, output and average per request) and the same figures by model, with a model_selection_mode of user_selected, auto or product_selected.
• Caveats: requests are counted on the server and are model calls, not prompts, so one prompt can count several times. Clients are identified by the integration ID they send, which GitHub calls a best-effort grouping, not a verified identity. Code review and cloud agent requests report product_selected, and code review models are reported as unknown.
• Access: enterprise owners, billing managers, organization owners and custom roles with the View Copilot Metrics permission, with the usage metrics policy enabled. The change is additive; existing fields are unchanged.
• GitHub notes this differs from agent app activity, which counts jobs started by agent apps such as Claude and Codex on GitHub.