GitHub adds CLI customization metrics to Copilot API
GitHub Copilot’s usage metrics API now reports how developers use agentic CLI customizations. The update gives enterprise and organization administrators new data on adoption and activity.

Broader coverage for Copilot CLI activity
GitHub has expanded the reporting coverage of its Copilot usage metrics API to include agentic customizations used through Copilot CLI. The new activity metrics cover skills, custom agents, Model Context Protocol (MCP) servers, slash commands and plugins.
The fields are available in enterprise and organization per-user reports, aggregate one-day reports, per-user 28-day reports and the day_totals entries in aggregate 28-day reports. In per-user reports, the metrics describe activity associated with one user. In aggregate reports, they describe activity across an enterprise or organization.
Reports identify popular customizations
The API adds totals_by_skill, totals_by_custom_agent, totals_by_mcp, totals_by_slash_cmd and totals_by_plugin arrays to show which items have the most recorded activity. Each array can list up to five items. Entries include an interaction_count, with the meaning of that count depending on the customization category.
For skills, slash commands and plugin skills, interaction_count measures invocations. For custom agents, it counts starts. For MCP servers, it counts connection attempts. These lists are intended to show which customizations are attracting the most activity, while the associated distinct-use fields provide a broader view of adoption.
Distinct-use data measures variety
Separate fields count the number of different items used: distinct_skill_use_count, distinct_custom_agent_use_count, distinct_mcp_use_count, distinct_slash_cmd_use_count and distinct_plugin_use_count. In a per-user report, each item used by that user counts once. In an aggregate report, each item used anywhere in the enterprise or organization counts once, rather than once per user.
The distinct-use counts include items that do not appear in the top-five arrays. GitHub says comparing these figures over time can show whether the range of customizations in use is expanding. The company also says administrators can use the data to identify which customizations are gaining traction, find enablement gaps and focus investment on automations developers value.
Privacy protections shape the labels
The reports display names for recognized GitHub-provided items, but customer-defined names are not shown. Skills, custom agents, MCP servers and plugins created by customers are grouped under the label other. Copilot CLI telemetry already groups customer-defined slash commands under custom, so reports use that label for those commands.
This means the metrics can describe activity patterns without exposing customer-defined customization names. Empty arrays and zero values indicate that no matching activity was recorded. When customization data is unavailable, the fields may be null or absent.
How MCP and plugin counts are calculated
MCP server activity follows a specific counting rule. The interaction_count increases when Copilot CLI attempts to connect or reconnect to a server. Both successful and failed attempts count. Repeatedly calling tools from a server that is already connected does not increase the count.
Plugin metrics are limited to skill invocations associated with a plugin. As a result, every plugin interaction also appears in the skill totals, while skill interactions that do not originate from a plugin appear only in the skill totals. GitHub cautions that plugin totals are a subset of skill totals and should not be added together.
Availability for administrators
GitHub says the reports are available to enterprise owners and billing managers, organization owners, and users with a custom organization or enterprise role that grants the View Copilot Metrics permission. The Copilot usage metrics policy must also be enabled.
The company directs administrators to the Copilot usage metrics API documentation to begin using the new fields. The expansion gives organizations a more detailed way to examine how agentic CLI features are being used while retaining grouping rules for customer-defined items.
Sources
- GitHub ChangelogAgentic CLI customizations now in the usage metrics API