官方更新 GitHub Changelog Thu, 17 Sep 2026 正文 ok 439 words

使用指标 API 现已支持 Agentic CLI 自定义项

英文标题:Agentic CLI customizations now in the usage metrics API

GitHub Copilot 在现有 CLI 使用报告中新增了 Agentic 活动指标,覆盖技能、自定义 Agent、Model Context Protocol(MCP)服务器、斜杠命令和插件。新字段已纳入企业和组织的按用户及聚合 1 天报告、按用户 28 天报告,以及聚合 28 天报告中的 `day_totals` 条目。管理员可据此识别最常使用的 CLI 自定义项、衡量不同自定义项的采用广度,并发现开发者赋能缺口。为保护隐私,客户自定义项名称不会在报告中展示,而是按 `other` 或 `custom` 分类汇总。
打开原文相关 8 · 机会 7
media16.2 KB media128.7 KB

中文

GitHub Copilot 扩展了现有 CLI 报告的覆盖范围,新增针对技能、自定义 Agent、Model Context Protocol(MCP)服务器、斜杠命令和插件的 Agentic 活动指标。

这些字段会出现在企业和组织的按用户及聚合 1 天报告、按用户 28 天报告,以及聚合 28 天报告中的 day_totals 条目中。

在按用户报告中,这些字段回答有关该用户活动的两个问题。在聚合报告中,它们回答整个企业或组织范围内的相同问题:

  • 哪些项目使用最多? totals_by_skill、totals_by_custom_agent、totals_by_mcp、totals_by_slash_cmd 和 totals_by_plugin 数组最多列出五个记录活动量最高的项目。每个条目均包含 interaction_count。视类别而定,该字段统计技能、斜杠命令和插件技能的调用次数,以及自定义 Agent 的启动次数和 MCP 服务器的连接尝试次数。
  • 使用了多少种不同的项目? distinct_skill_use_count、distinct_custom_agent_use_count、distinct_mcp_use_count、distinct_slash_cmd_use_count 和 distinct_plugin_use_count 字段统计所使用的不同项目数量。在按用户报告中,该用户使用过的每个项目计为一次。在聚合报告中,企业或组织内任意人员使用过的每个项目计为一次,而不是每位用户各计一次。这些计数包含排名前五以外的项目;对比其随时间的变化,可显示在用项目的多样性是否正在增长。

企业和组织管理员现在能够识别哪些 Copilot CLI 自定义项正在获得采用,发现赋能缺口,并将投入重点放在开发者认为有价值的自动化功能上。

  • 会显示已识别的 GitHub 提供项目的名称。为保护隐私,不显示客户自定义名称。技能、自定义 Agent、MCP 服务器和插件会归类至 other。Copilot CLI 遥测已将客户自定义斜杠命令归类至 custom,因此报告对斜杠命令使用该标签。
  • 对于 MCP 服务器,只有当 Copilot CLI 尝试连接或重新连接该服务器时,interaction_count 才会增加。成功和失败的尝试都会计入。从同一已连接服务器多次调用工具不会增加该计数。
  • 插件指标仅统计与插件关联的技能调用。因此,每次插件交互也会出现在技能总计中;但并非来自插件的技能交互只会出现在技能总计中。由于插件总计是技能总计的子集,不应将两者相加。
  • 空数组和零计数表示不存在匹配的活动。当自定义项数据不可用时,这些字段为 null 或不存在。
  • 企业所有者和计费管理员、组织所有者,以及拥有授予 View Copilot Metrics 权限的自定义组织或企业角色的任何人员均可访问这些报告。必须启用 Copilot 使用指标策略。

访问 Copilot usage metrics API 文档 以开始使用。

English / 原文

GitHub Copilot expands existing CLI report coverage with agentic activity metrics for skills, custom agents, Model Context Protocol (MCP) servers, slash commands, and plugins.

The fields appear in enterprise and organization per-user and aggregate 1-day reports, per-user 28-day reports, and the day_totals entries in aggregate 28-day reports.

In a per-user report, the fields answer two questions about that user’s activity. In an aggregate report, they answer the same questions across the enterprise or organization:

  • Which items are used most? The totals_by_skill, totals_by_custom_agent, totals_by_mcp, totals_by_slash_cmd, and totals_by_plugin arrays list up to five items with the most recorded activity. Each entry includes an interaction_count. Depending on the category, this counts invocations for skills, slash commands, and plugin skills, plus custom agent starts and MCP server connection attempts.
  • How many different items were used? The distinct_skill_use_count, distinct_custom_agent_use_count, distinct_mcp_use_count, distinct_slash_cmd_use_count, and distinct_plugin_use_count fields count the number of different items used. In a per-user report, each item that user used counts once. In an aggregate report, each item used by anyone in the enterprise or organization counts once, not once per user. These counts include items outside the top five, and comparing them over time shows whether the variety of items in use is growing.

Enterprise and organization administrators are now able to identify which Copilot CLI customizations are gaining traction, find enablement gaps, and focus investment on automations that developers find valuable.

  • Names for recognized GitHub-provided items are shown. To protect privacy, customer-defined names are not shown. Skills, custom agents, MCP servers, and plugins are grouped under other. Copilot CLI telemetry already groups customer-defined slash commands under custom, so reports use that label for slash commands.
  • For MCP servers, interaction_count increases only when Copilot CLI attempts to connect or reconnect to the server. Successful and failed attempts both count. Calling tools from the same connected server multiple times does not increase the count.
  • Plugin metrics count only skill invocations associated with a plugin. Every plugin interaction therefore also appears in the skill totals, but skill interactions that do not come from a plugin appear only in the skill totals. Because the plugin totals are a subset of the skill totals, you should not add the two together.
  • Empty arrays and zero counts indicate no matching activity. The fields are null or absent when customization data is unavailable.
  • 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.