Skip to main content
Glama
enesjakozh

MCP Prompts Server

by sparesparrow
npmGitHub
TypeScript
107
74
  • Apple
  • Linux
OverviewInspectNewEndpointsSchemaRelated ServersReviewsScore
Need Help?View Source CodeReport Issue

MCP Prompts Server

A robust, extensible server for managing, versioning, and serving prompts and templates for LLM applications, built on the Model Context Protocol (MCP).

🚀 Features

Dual Mode Operation

  • HTTP Mode - Traditional REST API server
  • MCP Mode - Model Context Protocol server for AI assistants

MCP Tools (7 Available)

  1. add_prompt - Add new prompts to collection
  2. get_prompt - Retrieve prompts by ID
  3. list_prompts - List prompts with filtering
  4. update_prompt - Update existing prompts
  5. delete_prompt - Remove prompts
  6. apply_template - Apply variables to template prompts
  7. get_stats - Get prompt statistics

Pre-loaded Templates

  • Code Review Assistant
  • Documentation Writer
  • Bug Analyzer
  • Architecture Reviewer
  • Test Case Generator

📦 Installation

npm install @sparesparrow/mcp-prompts

🎯 Quick Start

HTTP Mode

# Start HTTP server npm start # or MODE=http node dist/index.js # Server runs on http://localhost:3003 # API docs: http://localhost:3003/api-docs # Health check: http://localhost:3003/health

MCP Mode

# Start MCP server MODE=mcp node dist/index.js # or npm run start:mcp

🔧 Cursor Integration

  1. Configure Cursor MCP: Add to .cursor/mcp.json:
    { "mcpServers": { "mcp-prompts": { "command": "node", "args": ["dist/index.js"], "env": { "MODE": "mcp" } } } }
  2. Restart Cursor to load the MCP server
  3. Use in Cursor:
    • The MCP tools will be available in Cursor's AI assistant
    • You can ask Cursor to manage prompts using natural language

🐳 Docker Support

Build MCP Docker Image

pnpm run docker:build:mcp

Run with Docker Compose

pnpm run docker:up:mcp

View Logs

pnpm run docker:logs:mcp

📚 Documentation

🛠 Development

Prerequisites

  • Node.js >= 20.0.0
  • pnpm >= 9.0.0

Setup

# Install dependencies pnpm install # Build the project pnpm run build # Start development server pnpm run dev

Available Scripts

# Build pnpm run build pnpm run build:clean # Start servers pnpm run start # Default mode pnpm run start:http # HTTP mode pnpm run start:mcp # MCP mode # Docker pnpm run docker:build:mcp pnpm run docker:up:mcp pnpm run docker:down:mcp # Testing pnpm run test pnpm run test:watch # Linting & Formatting pnpm run lint pnpm run format

📊 Project Structure

mcp-prompts/ ├── src/ │ ├── index.ts # Main entry point │ ├── mcp-server.ts # MCP server implementation │ ├── http-server.ts # HTTP server implementation │ └── utils.ts # Utility functions ├── data/ │ └── sample-prompts.json # Pre-loaded templates ├── dist/ # Compiled output ├── docs/ # Documentation ├── examples/ # Usage examples └── docker/ # Docker configurations

🔍 Testing

MCP Server Testing

# Test MCP server functionality node test-mcp-complete.js

HTTP Server Testing

# Start HTTP server pnpm run start:http <<<<<<< HEAD ======= # Test endpoints curl http://localhost:3003/health curl http://localhost:3003/api-docs

📈 Usage Examples

Using MCP Tools

List All Prompts
{ "method": "tools/call", "params": { "name": "list_prompts", "arguments": {} } }

<<<<<<< HEAD

Build Issues

TypeScript Path Resolution Errors:

# Clear TypeScript cache rm -rf **/*.tsbuildinfo pnpm run clean pnpm run build

SWC Build Failures:

# Ensure SWC is installed pnpm add -D @swc/cli @swc/core # Clean and rebuild pnpm run clean pnpm run build

Workspace Dependency Issues:

# Reinstall workspace dependencies pnpm install --force pnpm run build

Missing Modules or Types:

  • Ensure you have built @mcp-prompts/core first
  • Check that all dist/ directories are up to date
  • If you change the shared config or move files, clean all dist/ directories and rebuild

Runtime Issues

Common Issues:

  • If you see errors about missing modules or types, ensure you have built @mcp-prompts/core first and that all dist/ directories are up to date.
  • If you change the shared config or move files, clean all dist/ directories and rebuild.

Architecture

Hexagonal Architecture (Ports & Adapters)

MCP Prompts follows a clean hexagonal architecture pattern:

  • Core: Pure domain logic without infrastructure dependencies
  • Ports: Interfaces defined in core package
  • Adapters: Infrastructure implementations in adapter packages
  • Apps: Composition and configuration in apps folder

Directory Structure

mcp-prompts/ ├── packages/ │ ├── core/ # Domain logic and ports │ ├── @sparesparrow/ # Shared packages │ └── adapters-*/ # Port implementations ├── apps/ │ └── server/ # MCP server application └── docs/ # Documentation

Development

Build Commands

# Build entire workspace pnpm run build # Development with watch mode pnpm run build:watch # Type checking pnpm run typecheck # Clean build artifacts pnpm run clean

Package-specific Commands

# Core package pnpm -F @sparesparrow/mcp-prompts-core build pnpm -F @sparesparrow/mcp-prompts-core test # Adapter packages pnpm -F @sparesparrow/mcp-prompts-adapters-file build pnpm -F @sparesparrow/mcp-prompts-adapters-mdc build # Server app pnpm -F apps/server build pnpm -F apps/server test

Testing

  • Vitest for unit tests
  • Playwright for e2e tests
  • Coverage > 90% for core packages
  • Integration tests for adapters

API Reference

For detailed API documentation, see:

References

Apply Template
{ "method": "tools/call", "params": { "name": "apply_template", "arguments": { "id": "code_review_assistant", "variables": { "language": "JavaScript", "code": "function hello() { console.log('Hello World'); }" } } } }
Add New Prompt
{ "method": "tools/call", "params": { "name": "add_prompt", "arguments": { "name": "My Custom Prompt", "content": "This is a custom prompt for {{subject}}", "isTemplate": true, "tags": ["custom", "example"], "variables": [ { "name": "subject", "description": "The subject to process", "required": true, "type": "string" } ] } } }

🌟 Features

  • Template Variables - Use {{variable}} syntax for dynamic content
  • Tag System - Organize prompts with tags for easy filtering
  • Metadata Support - Add categories, difficulty, time estimates
  • Version Control - Track prompt versions and changes
  • Error Handling - Comprehensive error handling and logging
  • TypeScript - Full TypeScript support with type definitions
  • Docker Ready - Containerized deployment support

🤝 Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Implement your changes
  4. Add tests if applicable
  5. Submit a pull request

📄 License

MIT License - see LICENSE file for details.

🆘 Support

Version: 3.0.8
Status: ✅ Production Ready
MCP Support: ✅ Full Implementation
Cursor Integration: ✅ Ready

Deploy Server
A
security – no known vulnerabilities
F
license - not found
A
quality - confirmed to work

How are these scores calculated?

hybrid server

The server is able to function both locally and remotely, depending on the configuration or use case.

단순화된 SOLID 아키텍처를 통해 프롬프트를 생성, 관리 및 템플릿화할 수 있어, 사용자는 범주별로 프롬프트를 구성하고 런타임에 템플릿을 채울 수 있습니다.

  1. 목차
    1. 특징
      1. 설치
        1. npx 사용(권장)
        2. 글로벌 설치
        3. Docker 사용
        4. 설치 확인
      2. 구성
        1. PostgreSQL 설정(STORAGE_TYPE=postgres인 경우 필수)
        2. MDC 설정(STORAGE_TYPE=mdc인 경우 필수)
      3. 용법
        1. Claude와 함께 사용
        2. 사용 가능한 도구
        3. API 사용 예
        4. 프롬프트 관리
        5. 워크플로에서 프롬프트 사용
      4. 프롬프트 형식
        1. 다양한 형식의 프롬프트 지원
          1. 형식 간 변환
          2. 템플릿 적용
          3. 변수 추출
          4. 다양한 형식으로 만들기
          5. 스토리지 어댑터와의 통합
        2. 스토리지 어댑터
          1. PostgreSQL 설정
        3. 도커 배포
          1. Docker Compose 오케스트레이션
        4. 개발
          1. 개발 워크플로
          2. 개발 명령
          3. 빌드 프로세스
          4. 테스트
          5. 디렉토리 구조
        5. 릴리스 프로세스
          1. 출시 전 체크리스트
          2. 버전 업데이트
          3. 출판
          4. 출시 후 검증
        6. 변경 사항
          1. [1.2.20] - 2025-03-14
          2. [1.2.19] - 2024-03-16
          3. [1.2.18] - 2024-03-14
          4. [1.2.0] - 2025-03-14
          5. [1.1.0] - 2024-03-01
          6. [1.0.0] - 2024-02-15
        7. 모범 사례
          1. 특허

            Related Resources

            Related MCP Servers

            • TypeScript Prompt MCP Server

              gpaul-mcp
              A
              security
              A
              license
              A
              quality
              Provides pre-defined prompt templates for AI assistants to generate comprehensive plans for TypeScript projects, API architectures, and GitHub workflows.
              Last updated -
              2
              1
              MIT License

            • PromptLab MCP Server

              iRahulPandey
              -
              security
              A
              license
              -
              quality
              Serves prompt templates through a standardized protocol for transforming basic user queries into optimized prompts for AI systems.
              Last updated -
              6
              Apache 2.0

            • MLflow Prompt Registry MCP Server

              B-Step62
              -
              security
              F
              license
              -
              quality
              Enables access to prompt templates managed in MLflow through Claude Desktop, allowing users to instruct Claude with saved templates for repetitive tasks or common workflows.
              Last updated -
              2

            • Prompts MCP Server

              tanker327
              A
              security
              A
              license
              A
              quality
              A Model Context Protocol server for managing prompt templates as markdown files with YAML frontmatter, allowing users and LLMs to easily add, retrieve, and manage prompts.
              Last updated -
              5
              3
              9
              MIT License
              • Apple

            View all related MCP servers

            Appeared in Searches

            New MCP Servers

            MCP directory API

            We provide all the information about MCP servers via our MCP API.

            curl -X GET 'https://glama.ai/api/mcp/v1/servers/sparesparrow/mcp-prompts'

            If you have feedback or need assistance with the MCP directory API, please join our Discord server