PipeCD Docs MCP Server
Enables searching and retrieving official PipeCD documentation through full-text search and document retrieval. Clones PipeCD docs from GitHub and indexes Markdown files for quick access to titles and content.
README
PipeCD Docs MCP Server
A local MCP server for searching official PipeCD docs.
Overview
This server clones the official PipeCD docs from GitHub and provides simple full-text search and document retrieval APIs via the MCP protocol.
Documentation is cloned into a temporary directory, and Markdown files are indexed by extracting their titles and content.
Usage
Prerequisites
-
Create a PAT with
read:packagesscope. Details -
Add the following config to your
.npmrcfile.
@pipe-cd:registry=https://npm.pkg.github.com
//npm.pkg.github.com/:_authToken=YOUR_TOKEN
Cursor
Quick Install with Cursor Deeplink
For Cursor users, you can install the MCP server with a single click using the deeplink below:
This will automatically configure the MCP server in your Cursor settings. After clicking the link, Cursor will prompt you to install the server.
Install Using npx
Add the following config to your mcp.json.
{
"mcpServers": {
"pipe-cd.docs-mcp-server": {
"type": "stdio",
"command": "npx",
"args": [
"@pipe-cd/docs-mcp-server@latest"
],
}
}
}
<!-- TODO: Add Claude Desktop etc. -->
Tools
search_docs
Executes a full-text search on PipeCD docs.
- Parameters:
query: Search keywords (space-separated, AND search)offset: Start position for resultslimit: Number of results to return (default: 20)
read_docs
Returns the content of the specified page.
- Parameters:
path: Relative path of the document (after "docs/content/en/")
Implementation Notes
- Uses sparse-checkout to minimize clone size and speed up the process.
- Titles are extracted from the Markdown front matter.
- The search logic of
search_docsis so simple for now.
Contributing
Code of Conduct
PipeCD follows the CNCF Code of Conduct. Please read it to understand which actions are acceptable and which are not.
Get Involved
- Slack:
#pipecdchannel on CNCF Slack for discussions related to PipeCD development. - Community Meeting: Every other Wednesdays. Search here.
Improvements
- Bug:
- Please open an Issue and describe the problem. Or, open a PR with if it's easy one.
- Enhancement / Feature Request:
- For small changes including docs or adding tests, please open a PR.
- For new features or big enhancement, basically, please open an Issue and discuss there before sending a PR. We cannot accept all requests in some cases.
- Security issue:
- Send an email to the core maintainers. DO NOT report on Issues.
Release Flow
- Run the
prepare releaseworkflow with specifing the new version. It will create a PR to update the version. - Review the PR and merge it.
- Create a Release on GitHub with a new tag.
Then, the
release packageworkflow will publish a new npm package.
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。