
Yak MCP
Enables coding agents to speak aloud using text-to-speech functionality. Works with agents running inside devcontainers and provides configurable voice settings for creating chatty AI companions.
README
yak-mcp
<p align="center"> <img width="400" alt="yak-mcp" src="./yak-mcp.png" /> </p>
<p align="center">An MCP that turns coding agents into "yak (chatty)" companions.</p>
<p align="center">This is a typical voice-notifying MCP, but it also works with coding agents running inside a devcontainer.</p>
Installation
Download the installer for your platform (Windows, Linux) from the Release Page.
Note for macOS users: I was unable to build a macOS installer in my environment. If you need it, please build it yourself following the instructions in this repository. Thank you for your understanding.
Setup
When you first launch the application, the settings screen will appear.
MCP Server
- Host: Enter the IP address to bind the server.
- If you want to use it from a coding agent inside a devcontainer, setting it to
0.0.0.0
is recommended.
- If you want to use it from a coding agent inside a devcontainer, setting it to
- Port: Specify an available port for the server.
TTS (Text-to-Speech)
- Lang: Choose the language for speech output.
- Volume: Adjust the speaking volume.
- Pitch: Set the pitch (voice height).
- Rate: Set the reading speed.
After saving your settings, the MCP server will start on the specified port.
From the next launch onwards, the MCP server will automatically start when the app launches.
You can reopen the settings screen from the task tray whenever you want to change your configuration.
Configuring Your Coding Agent
Add the following configuration to your coding agent to connect to the MCP server:
"yak": {
"type": "http",
"url": "http://localhost:8080/mcp"
}
- If you are using Docker Desktop and want to connect from a devcontainer, set the host to
host.docker.internal
. - If you are connecting from a local coding agent,
localhost
works fine. - Make sure the port number matches the one specified in the settings.
To allow unconditional access to this tool, enable the yak.say
permission.
Testing
Once the setup is complete, start your coding agent and try the following instruction:
use yak, say "Hello."
If the agent speaks, the setup is successful.
Enjoy using your noisy, talkative coding agent!
For Developers
Recommended IDE Setup
Project Setup
Install
pnpm install
Development
pnpm dev
Build
# For windows
$ pnpm build:win
# For macOS
$ pnpm build:mac
# For Linux
$ pnpm build:linux
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。