covate

Overview概述

covate is open-core developer learning. A free open-source MCP sidecar quizzes you on your AI-assisted code changes when you ask and keeps persistent debugging memory across sessions, plus a cloud learning platform you sign into with GitHub — it keeps a record of your sessions. It plugs into AI coding assistants rather than replacing them.covate 是开源核心的开发者学习。一个免费的开源 MCP 边车会在你提出请求时，就你借助 AI 完成的代码改动向你提问，并在多次会话之间保留持久的调试记忆；此外还有一个云端学习平台，你用 GitHub 登录——它会保存你的会话记录。它接入 AI 编程助手，而不是取代它们。

Who it’s for适合谁

Developers who use AI coding assistants and want to actually retain what the AI does with them — turning passive acceptance into durable learning.适合使用 AI 编程助手、并希望真正记住 AI 与自己一起做了什么的开发者——把被动接受变成持久的学习。

What it does它能做什么

• Quizzes you on your own AI-assisted changes when you ask在你提出请求时，就你自己借助 AI 完成的改动向你提问
• Persistent debugging memory across sessions跨会话持久保留的调试记忆
• MCP sidecar that plugs into AI coding assistants接入 AI 编程助手的 MCP 边车
• Open-core: free sidecar plus a cloud platform you sign into with GitHub开源核心：免费的边车，外加一个用 GitHub 登录的云端平台

How it works如何运作

1. It quizzes you on the diff它就 diff 向你提问

   Covate is an MCP server that sits beside your AI coding assistant. When the assistant writes something, you can ask for a quiz on it — logic, security or performance implications — so accepting a change and understanding it stop being the same action.Covate 是一个伴随在你的 AI 编程助手旁边的 MCP 服务器。当助手写出一些东西时，你可以请求就它出一份测验——逻辑、安全或性能方面的影响——于是接受一处改动和理解它不再是同一个动作。

2. And it remembers debugging而且它记得调试

   debug_record keeps a per-project debugging memory the assistant can look up when a similar error appears, instead of solving the same problem twice.debug_record 保留一份按项目划分的调试记忆，当类似的错误出现时，助手可以查找它，而不必把同一个问题解两遍。

3. A free learning ledger一个免费的学习账本

   A hosted learning ledger at covate.org shows your sessions, your accuracy and the topics you answer worst.covate.org 上的一个托管学习账本会显示你的会话、你的准确率，以及你答得最差的主题。

Built with技术构成

A Python (3.11+) MCP server, MIT-licensed, installed from its GitHub repository (not on npm or PyPI). The learning ledger at covate.org is a separate web app.一个 Python（3.11+）MCP 服务器，采用 MIT 许可，从其 GitHub 仓库安装（不在 npm 或 PyPI 上）。covate.org 上的学习账本是一个独立的网页应用。

Current status当前状态

Live at covate.org; the MCP is open source on GitHub.已在 covate.org 上线；MCP 在 GitHub 上开源。

Setup安装配置

covate is a Python MCP server (3.11+) that installs from its GitHub repository — it is not published to npm or PyPI, so neither npx covate nor pip install covate will find it. The installer clones the repo to ~/covate, builds a virtualenv beside it, and prints the executable path you point your assistant at.covate 是一个 Python MCP 服务器（3.11+），从其 GitHub 仓库安装——它没有发布到 npm 或 PyPI，所以 npx covate 和 pip install covate 都找不到它。安装程序会把仓库克隆到 ~/covate，在旁边构建一个 virtualenv，并打印出你要让助手指向的可执行文件路径。

1. Install — macOS / Linux安装 — macOS / Linux

   Detects uv, conda, or venv; clones to ~/covate; installs dependencies; prints the command path to use below.检测 uv、conda 或 venv；克隆到 ~/covate；安装依赖；打印出下面要用的命令路径。

   curl -fsSL https://raw.githubusercontent.com/SunflowersLwtech/covate/main/scripts/install.sh | bash

2. Install — Windows (PowerShell)安装 — Windows（PowerShell）

   irm https://raw.githubusercontent.com/SunflowersLwtech/covate/main/scripts/install.ps1 | iex

3. Claude Code — one commandClaude Code — 一条命令

   Use the path the installer printed. On a default install that is ~/covate/covate/bin/covate (Windows: %USERPROFILE%\covate\covate\Scripts\covate.exe).

   claude mcp add covate -- ~/covate/covate/bin/covate

4. Claude Code — config file (~/.claude.json)Claude Code — 配置文件（~/.claude.json）

   Equivalent to the command above if you would rather edit JSON. Transport is stdio, which is the default.如果你更愿意编辑 JSON，这与上面的命令等价。传输方式是 stdio，这也是默认值。

   {
     "mcpServers": {
       "covate": {
         "command": "~/covate/covate/bin/covate"
       }
     }
   }

5. Cursor — ~/.cursor/mcp.json

   Settings → MCP → Add Server writes the same entry. Windsurf takes an identical block in ~/.codeium/windsurf/mcp_config.json. Restart the editor afterwards.设置 → MCP → 添加服务器会写入同样的条目。Windsurf 在 ~/.codeium/windsurf/mcp_config.json 里使用一个完全相同的块。之后重启编辑器。

   {
     "mcpServers": {
       "covate": {
         "command": "~/covate/covate/bin/covate"
       }
     }
   }

6. Optional — sync sessions to the learning ledger可选 — 将会话同步到学习账本

   Sign in at covate.org with GitHub, reveal your sync token in the dashboard, then run this from a project directory. The MCP works fully without it.用 GitHub 在 covate.org 登录，在仪表盘里显示你的同步令牌，然后在一个项目目录里运行下面这条命令。不这样做，MCP 也能完整工作。

   COVATE_SYNC_URL=https://covate.org COVATE_SYNC_TOKEN=<your token> ~/covate/covate/bin/python -m covate.platform_sync

Environment variables环境变量

• MCP_DEBUG default: false Optional. Debug logging; accepts true, 1, yes, or on.可选。调试日志；接受 true、1、yes 或 on。
• COVATE_SYNC_URL default: unset Optional. Set to https://covate.org only when pushing sessions to the hosted ledger.可选。仅在将会话推送到托管账本时设为 https://covate.org。
• COVATE_SYNC_TOKEN default: unset Optional. Your per-user token from covate.org/dashboard → Sync token.可选。你在 covate.org/dashboard → Sync token 处取得的个人令牌。

If it doesn’t work如果它不工作

Commands come from the covate repository — github.com/SunflowersLwtech/covate — and are current as of its README and scripts/install.sh. If the repo and this page ever disagree, the repo is right.这些命令来自 covate 仓库——github.com/SunflowersLwtech/covate——并以截至其 README 和 scripts/install.sh 的内容为准。如果仓库和本页有任何不一致，以仓库为准。

Pricing价格

Free.免费。

• The MCP server is free and openMCP 服务器免费且开放
• The hosted learning ledger is free托管的学习账本是免费的

FAQ常见问题