// AI Platform Drift
GitHub Copilot usage API now reports custom skill and agent names
GitHub Copilot changelog · published 2026-10-23 04:58 UTC
WHAT CHANGEDThe Copilot usage metrics API added five new fields (totals_by_skill, totals_by_custom_agent, totals_by_mcp, totals_by_slash_cmd, totals_by_plugin) that show configured names of customer-defined items instead of grouping them as 'other' or 'custom'. Each list now includes up to 25 entries per record, increased from five. Names appear only for report days processed after 22 October 2026. Copilot CLI 1.0.97+ and SDK v1.0.21+ are required for name reporting.
WHY IT MATTERSExisting API consumers parsing usage reports will see new fields in the response schema and different values in existing aggregation arrays. Dashboards or scripts that expect 'other' or 'custom' labels will now receive actual configured names for internal customizations. Historical reports remain unchanged, creating a data discontinuity at the release date.
WHAT TO DOReview code that parses Copilot usage API responses for hardcoded field expectations or 'other'/'custom' string matching. Update Copilot CLI to 1.0.97+ and SDK to v1.0.21+ if you want name-level visibility. Verify that configured names for skills, agents, and MCP servers are appropriate for report distribution outside your organization, as they are reported exactly as configured (trimmed to 128 characters).
Model-reported confidence 85% (not measured accuracy)
Primary source ↗
written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
+## Custom skills, agents, and MCP servers in usage metrics API +https://github.blog/changelog/2026-10-22-custom-skills-agents-and-mcp-servers-in-usage-metrics-api +Thu, 22 Oct 2026 22:19:17 +0000 + +Copilot usage metrics now show the names of the skills, custom agents, Model Context Protocol (MCP) servers, slash commands, and plugins that your organization creates, instead of grouping them under other or custom. This builds on the agentic CLI customization metrics released in September. +What’s new + +Named entries: In enterprise and organization reports, including per-user 1-day and 28-day reports, aggregated 1-day reports, and day_totals in aggregated 28-day reports, the totals_by_skill, totals_by_custom_agent, totals_by_mcp, totals_by_slash_cmd, and totals_by_plugin arrays now show the configured name of a customer-defined item instead of other or custom. +More entries: Each totals_by_* list now includes up to 25 entries per record, up from five, so you can see more of the long tail of customizations your developers use. Any remaining activity is combined into a trailing other entry. +Unchanged counts: interaction_count and the distinct_*_use_count fields are calculated the same way as before. + +Why this matters +The customizations that matter most to an organization are often the ones its own teams build. Administrators can now see which internal skills, custom agents, and MCP servers developers actually use, track adoption after rolling out a new one, and focus enablement where it is needed. +Important notes + +Coverage: Names are reported from Copilot CLI (interactive sessions and prompt mode), the GitHub Copilot app, Agent Client Protocol (ACP) integrations such as JetBrains, and the Copilot SDK. Prompt mode, the app, and ACP integrations require Copilot CLI 1.0.97 or later. The Copilot SDK requires v1.0.21 or later. IDEs report names when their agent experiences run on the Copilot SDK v1.0.21 or later. Names aren’t reported from an IDE’s built-in chat telemetry. +Configured names: Names are reported as they are configured, trimmed and limited to 128 characters. Names that look like file paths stay grouped under other or custom. Review names before sharing reports outside your organization. +Accuracy: These metrics come from client telemetry and can undercount activity, for example when clients are outdated or telemetry is blocked. Use them to understand trends and adoption rather than for exact accounting. +History: Names appear starting with report days processed after this release. Earlier days still show other or custom. +Access: Reports are available to enterprise owners and billing managers, organization owners, and anyone with a custom organization or enterprise role that grants the “View Copilot Metrics” permission. The Copilot usage metrics policy must be enabled. + +Visit the Copilot usage metrics API documentation to get started. + +The post Custom skills, agents, and MCP servers in usage metrics API appeared first on The GitHub Blog.
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
| triage verdict | MATERIAL at 0.92, needed 0.5 |
|---|---|
| triage reasoning | Usage metrics API schema expanded with new named fields and increased capacity; version requirements introduced. |
| signals | API schema change: new fields added (totals_by_skill, totals_by_custom_agent, totals_by_mcp, totals_by_slash_cmd, totals_by_plugin) · list capacity increased: 5 entries → 25 entries per record · behavioral change: named entries replace 'other' or 'custom' grouping · version floor introduced: Copilot CLI 1.0.97+, SDK v1.0.21+ · retroactive behavior: names only appear for report days processed after release |
| suppression | — |
| triage model | anthropic/claude-haiku-4.5 ($0.00000) |
| writer model | anthropic/claude-sonnet-4.5 ($0.00000) |