Dagger MCP Server
An MCP (Model Context Protocol) server that exposes Dagger CI/CD capabilities as tools for AI assistants.
接入你的智能体
从发布者处安装。不是 PlugMCP 端点。
远程 URL 还是本机命令? 两种安装方式的区别
来自发布者
An MCP (Model Context Protocol) server that exposes Dagger CI/CD capabilities as tools for AI assistants.
MCP server exposing Dagger CI/CD capabilities as tools for AI assistants
| Tool |
|---|
| Description |
|---|
dagger_version | Get the installed Dagger version |
dagger_call | Run a Dagger function from a module |
dagger_functions | List all functions in a Dagger module |
dagger_build | Build a container image using Dagger |
dagger_query | Execute a Dagger GraphQL query |
dagger_shell | Run dagger shell with a command |
dagger_init | Initialize a new Dagger module |
dagger_go | Run Dagger with a Go module |
git clone https://github.com/user/dagger-mcp-server.git
cd dagger-mcp-server
cargo install --path .
Add to ~/.config/opencode/opencode.json:
{
"mcpServers": {
"dagger": {
"type": "local",
"command": ["dagger-mcp-server"],
"enabled": true
}
}
}
Add to Claude Desktop config:
{
"mcpServers": {
"dagger": {
"command": "dagger-mcp-server"
}
}
}
Add to Cursor MCP settings:
{
"mcpServers": {
"dagger": {
"command": "dagger-mcp-server"
}
}
}
Once configured, your AI assistant can:
# Check Dagger version
"What version of Dagger is installed?"
# List functions in a module
"Show me the available functions in ./my-dagger-module"
# Run a Dagger function
"Run the 'build' function in the ./ci module"
# Build and push an image
"Build using the 'container' function and push to ghcr.io/myorg/myapp:v1"
# Build
cargo build
# Run locally (stdio mode)
cargo run
# Test with Dagger
dagger version
cargo run
MIT
MCP server exposing Dagger CI/CD capabilities as tools for AI assistants
工具
暂无工具列表。远程服务器在首次健康检查后会获得工具列表。
常见问题
在本页复制适用于 Cursor、Claude Desktop、Claude Code 或 VS Code 的 stdio 配置。它使用发布者的安装方式,而不是 PlugMCP 代理。
此记录未列出必需的环境变量。
Dagger 通过 stdio 在本地运行。
上次检查时间:2026-10-27。 MCP server exposing Dagger CI/CD capabilities as tools for AI assistants