Hacktoberfest 2026:维护者为十月标记出来的 issue,仍然开放、适合新手。 浏览 Hacktoberfest issue

Per-tool latency histograms and error-rate view

未关闭
#984 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
35/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
活跃
技术栈
typescript

调研方向

Start from the tool_invocations table and the existing usage view, which already hold per-call duration and status; the audit log is the other source. Find how the usage view queries that data, then write an aggregate query for calls, error rate, p50/p95/p99 and last error per tool over a time range defaulting to 24 h. Done when a per-server and per-connector view shows these figures and the range selector works. Scope is open, and #990 overlaps, so confirm the plan with a maintainer first.

由索引模型根据 Issue 内容生成。

描述

help wanted roadmap

Every tool call is already written to the audit log with duration and status, and there is a tool_invocations table behind the usage view. What is missing is the read side: a per-tool p50/p95/p99 and error rate over a time range, so an operator can see which connector is slow or failing without writing SQL.

Scope

  • Aggregate query over the existing invocation data — no new write path.
  • A view per MCP server and per connector: calls, error rate, p50/p95/p99, last error message.
  • Time range selector; default last 24 h.

Related: #990 (OpenTelemetry exporter) is the other half of this, for people who already have a metrics stack.

主要语言
TypeScript
星标
856
派生
82
平均合并
6 小时 44 分钟
30 天内合并 PR
181

环境准备

这个项目没有提供开发容器、Dockerfile 或贡献指南,环境需要你自己搭建:先看它的 README,通用步骤见我们的新手贡献指南。

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

HelpCode-ai/anythingmcp 的其他 Issue

查看 HelpCode-ai/anythingmcp 的全部 Issue

相似的 Issue

更多 TypeScript Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。