MCP Project Helper
A lightweight, extensible MCP server for running prompt-based tools and file utilities, with support for custom prompts.
mcp-project-helper
A lightweight, extensible MCP (Model Context Protocol) server for running prompt-based tools and file utilities. Designed for easy integration, testing, and extension with custom prompts.
Features
- Prompt-based tools: Easily add new tools by writing simple JSON prompt files.
- File utilities: Includes tools for reading, writing, moving, and deleting files and directories.
- Custom prompts: Place your own prompt definitions in the
custom_prompts/directory. - Multiple transports: Supports STDIO, SSE, and HTTP for flexible integration.
- Extensive tests: Includes a test script to verify all tool endpoints.
Getting Started
Build Locally
make build-local
⚡ Quick Start
Install via go install
To quickly install the latest version from the repository:
go install github.com/ad/mcp-project-helper@latest
The binary will appear in $GOBIN or $HOME/go/bin (make sure this path is in your $PATH).
1. Build from source
# Clone the repository
git clone https://github.com/ad/mcp-project-helper.git
cd mcp-project-helper
go mod tidy
# Local build
make build-local
# Or manually
go build -o mcp-project-helper main.go
# Local build
make build-local
# Or manually
go build -o mcp-project-helper main.go
# Docker build
make build
Run the Server
- STDIO (default):
./mcp-project-helper
- SSE:
./mcp-project-helper -transport sse -port 8080 - HTTP:
./mcp-project-helper -transport http -port 8080
Run Tests
./test.sh
🔌 Integration
VS Code
go install github.com/ad/mcp-project-helper@latest
Добавьте в settings.json:
{
"mcp": {
"servers": {
"helper": {
"type": "stdio",
"command": "/absolute/path/to/project-helper",
"args": ["-transport", "stdio"]
}
}
}
}
Docker (VS Code)
{
"mcp": {
"servers": {
"helper": {
"type": "stdio",
"command": "docker",
"args": [
"run", "--rm", "-i",
"danielapatin/mcp-project-helper:latest",
"-transport", "stdio"
]
}
}
}
}
Claude Desktop
{
"mcpServers": {
"helper": {
"command": "/absolute/path/to/mcp-project-helper",
"args": ["-transport", "stdio"]
}
}
}
Adding Custom Tools
- Create a JSON file in
custom_prompts/(seepalette.jsonfor an example). - Each tool must define a
descriptionand apromptfield. - The tool will be automatically registered and available via the MCP protocol.
Example Tools
- tool-generator: Generates a tool description and prompt template based on a user query.
- palette: Suggests a harmonious color palette for a given color.
Project Structure
main.go— Main server entry pointprompts/— Built-in prompt toolscustom_prompts/— User-defined prompt toolstest.sh— End-to-end test scriptMakefile— Build and run commands
License
MIT
Похожие серверы
Scout Monitoring MCP
спонсорPut performance and error data directly in the hands of your AI assistant.
Alpha Vantage MCP Server
спонсорAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
Buildable
Official MCP server for Buildable AI-powered development platform. Enables AI assistants to manage tasks, track progress, get project context, and collaborate with humans on software projects.
Code Sync MCP Server
Hot reload remote containerized Python applications directly from your IDE.
ask-gemini-mcp
MCP server that enables AI assistants to interact with Google Gemini CLI
Placid.app
Generate images and videos from templates using the Placid.app API.
Chrome Debug MCP Server
Automate your browser by connecting to Chrome's debugging port, preserving your login state.
MCP All Servers
A collection of reference implementations for the Model Context Protocol (MCP), showcasing servers built with TypeScript and Python SDKs.
Learn MCP
A sample project for learning MCP development, featuring a calculator for math operations and a prompt flow server for various templates.
MCP Gateway
A reverse proxy gateway for managing and accessing multiple MCP servers through a single entry point, deployable via Docker.
EdgeOne Pages MCP
An MCP server and client implementation for EdgeOne Pages Functions, supporting OpenAI-formatted requests.
MCP Bridge for Zotero
MCP server that enables AI assistants to build, test, and debug Zotero plugins via 26 tools for UI inspection, JS execution, logging, and more.