webdav-mcp

webdav-mcp

Provides 8 WebDAV file operations as MCP tools for AI agents to interact with remote WebDAV servers.

Category
访问服务器

README

webdav-mcp

A zero-dependency WebDAV MCP server — single static binary, no Node.js or Python required.

Implements the Model Context Protocol (MCP) over JSON-RPC 2.0 via stdio, exposing 8 WebDAV operations as tools that any MCP-compatible AI agent can call.

Why this exists

Every other WebDAV MCP server on GitHub requires a Node.js or Python runtime. This one is a single self-contained binary (~100–160 KB) compiled from Zig. It runs anywhere without installing a runtime, and it cross-compiles to all major platforms from one machine.

Tools

Tool Description
list List files and directories at a path
read Read file contents as text
write Write (create or overwrite) a file
delete Delete a file or directory
mkdir Create a directory
move Move or rename a file or directory
copy Copy a file or directory (sends Depth: infinity per RFC 4918 §9.8.2)
stat Get metadata for a single file or directory (PROPFIND Depth:0)

Tool parameters

list

  • path (required) — directory path relative to WebDAV root
  • recursive (boolean, default false) — if true, lists recursively using Depth: infinity. Note: some servers (IIS, some nginx configs) block recursive listing.

read

  • path (required) — file path relative to WebDAV root
  • max_bytes (integer, default 1048576) — maximum file size to read in bytes. Returns an error if the file is larger. Increase this for large files.

write

  • path (required) — file path relative to WebDAV root
  • content (required) — content to write
  • content_type (string, default application/octet-stream) — MIME type for the file (e.g. text/plain, application/json)
  • create_parents (boolean, default false) — if true and the parent directory does not exist, automatically creates all missing ancestor directories via MKCOL before writing

stat

  • path (required) — path to stat relative to WebDAV root
  • Returns: file or dir or missing, plus size=, modified=, etag=, and type= when available

Configuration

Set these environment variables before launching the server:

Variable Description Example
WEBDAV_URL Base URL of the WebDAV server http://192.168.1.10:8080
WEBDAV_USER Username admin
WEBDAV_PASS Password secret

Credential security

Credentials are sent via native Zig HTTP networking using the Authorization: Basic ... header. No external curl process is spawned.

Building from source

Requires Zig 0.16.0.

# Development build
zig build

# Optimized release build (~100-160 KB)
zig build -Doptimize=ReleaseSmall

# Cross-compile for all platforms (outputs to zig-out/release/)
zig build cross

Cross-compilation targets

File Platform
webdav-mcp-macos-arm64 macOS Apple Silicon
webdav-mcp-macos-x86_64 macOS Intel
webdav-mcp-linux-arm64 Linux ARM64 (static musl)
webdav-mcp-linux-x86_64 Linux x86-64 (static musl)
webdav-mcp-linux-armv7 Linux ARMv7 / Raspberry Pi (static musl)
webdav-mcp-windows-x86_64.exe Windows x86-64

Integration with nullclaw / *Claw bots

Add an mcp_servers entry to your bot's config.json:

{
  "mcp_servers": {
    "webdav": {
      "command": "/path/to/webdav-mcp",
      "args": [],
      "env": {
        "WEBDAV_URL": "http://your-webdav-server:8080",
        "WEBDAV_USER": "your-username",
        "WEBDAV_PASS": "your-password"
      }
    }
  }
}

The server speaks JSON-RPC 2.0 over stdio (newline-delimited), which is the standard MCP transport. Once registered, the agent can call all 8 tools as normal tool calls.

Running tests

zig build test --summary all

All unit tests cover XML parsing, JSON-RPC dispatch, argument extraction, URL construction, percent-encoding, entity decoding, and HTTP method routing.

License

MIT — see LICENSE.

<!-- mcp-name: io.github.vernonstinebaker/webdav-mcp -->

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选