• 🔥 Intelligent Hot-Reload System → Update prompts instantly without restarts
• 🎯 Advanced Template Engine → Nunjucks-powered with conditionals, loops, and dynamic data
• ⚡ Multi-Phase Orchestration → Robust startup sequence with comprehensive health monitoring
• 🚀 Universal MCP Compatibility → Works flawlessly with Claude Desktop, Cursor Windsurf, and any MCP client
• 🧠 Prompt Chain Workflows → Build complex multi-step AI processes
• 📊 Real-Time Diagnostics → Performance metrics and health validation built-in

Transform your AI assistant experience from scattered prompts to a powerful, organized command library that works across any MCP-compatible platform.

🎯 The Future is Here: Manage Your AI's Capabilities FROM WITHIN the AI Conversation

This isn't just another prompt server – it's a living, breathing prompt ecosystem that evolves through natural conversation with your AI assistant. Imagine being able to:

# 🗣️ Create new prompts by talking to your AI
"Hey Claude, create a new prompt called 'code_reviewer' that analyzes code for security issues"
→ Claude creates, tests, and registers the prompt instantly

# ✏️ Refine prompts through conversation
"That code reviewer prompt needs to also check for performance issues"
→ Claude modifies the prompt and hot-reloads it immediately

# 🔍 Discover and iterate on your prompt library
>>listprompts
→ Browse your growing collection, then ask: "Improve the research_assistant prompt to be more thorough"

🌟 Why This Changes Everything:

• 🔄 Self-Evolving System: Your AI assistant literally builds and improves its own capabilities in real-time
• 🎮 No Context Switching: Never leave your AI conversation to manage prompts – everything happens inline
• 🧠 Collaborative Intelligence: You and your AI work together to craft the perfect prompt library
• ⚡ Instant Gratification: Create → Test → Refine → Deploy in seconds, not minutes
• 🌱 Organic Growth: Your prompt library naturally evolves based on your actual usage patterns

This is what conversational AI infrastructure looks like – where the boundary between using AI and building AI capabilities disappears entirely.

Get your AI command center running in under a minute:

# Clone → Install → Launch → Profit! 🚀
git clone https://github.com/minipuft/claude-prompts-mcp.git
cd claude-prompts-mcp/server && npm install && npm run build && npm start

Drop this into your claude_desktop_config.json:

{
  "mcpServers": {
    "claude-prompts-mcp": {
      "command": "node",
      "args": ["E:\\path\\to\\claude-prompts-mcp\\server\\dist\\index.js"],
      "env": {
        "MCP_PROMPTS_CONFIG_PATH": "E:\\path\\to\\claude-prompts-mcp\\server\\promptsConfig.json"
      }
    }
  }
}

Configure your MCP client to connect via STDIO transport:

• Command: node
• Args: ["path/to/claude-prompts-mcp/server/dist/index.js"]
• Environment: MCP_PROMPTS_CONFIG_PATH=path/to/promptsConfig.json

💡 Pro Tip: Use absolute paths for bulletproof integration across all MCP clients!

Your AI command arsenal is ready, and it grows through conversation:

# Discover your new superpowers
>>listprompts

# Execute lightning-fast prompts
>>friendly_greeting name="Developer"

# 🚀 NEW: Create prompts by talking to your AI
"Create a prompt called 'bug_analyzer' that helps me debug code issues systematically"
→ Your AI creates, tests, and registers the prompt instantly!

# 🔄 Refine prompts through conversation
"Make the bug_analyzer prompt also suggest performance improvements"
→ Prompt updated and hot-reloaded automatically

# Handle complex scenarios with JSON
>>research_prompt {"topic": "AI trends", "depth": "comprehensive", "format": "executive summary"}

# 🧠 Build your custom AI toolkit naturally
"I need a prompt for writing technical documentation"
→ "The bug_analyzer needs to also check for security issues"
→ "Create a prompt chain that reviews code, tests it, then documents it"

🌟 The Magic: Your prompt library becomes a living extension of your workflow, growing and adapting as you work with your AI assistant.

# Edit any prompt file → Server detects → Reloads automatically → Zero downtime

Analyze {{content}} for {% if focus_area %}{{focus_area}}{% else %}general{% endif %} insights.

{% for requirement in requirements %}
- Consider: {{requirement}}
{% endfor %}

{% if previous_context %}
Build upon: {{previous_context}}
{% endif %}

Phase 1: Foundation → Config, logging, core services
Phase 2: Data Loading → Prompts, categories, validation
Phase 3: Module Init → Tools, executors, managers
Phase 4: Server Launch → Transport, API, diagnostics

{
  "id": "content_analysis_chain",
  "name": "Content Analysis Chain",
  "isChain": true,
  "chainSteps": [
    {
      "stepName": "Extract Key Points",
      "promptId": "extract_key_points",
      "inputMapping": { "content": "original_content" },
      "outputMapping": { "key_points": "extracted_points" }
    },
    {
      "stepName": "Analyze Sentiment",
      "promptId": "sentiment_analysis",
      "inputMapping": { "text": "extracted_points" },
      "outputMapping": { "sentiment": "analysis_result" }
    }
  ]
}

graph TB
    A[Claude Desktop] -->|MCP Protocol| B[Transport Layer]
    B --> C[🧠 Orchestration Engine]
    C --> D[📝 Prompt Manager]
    C --> E[🛠️ MCP Tools Manager]
    C --> F[⚙️ Config Manager]
    D --> G[🎨 Template Engine]
    E --> H[🔧 Management Tools]
    F --> I[🔥 Hot Reload System]

    style C fill:#ff6b35
    style D fill:#00ff88
    style E fill:#0066cc

This server implements the Model Context Protocol (MCP) standard and works with any compatible client:

💡 Developer Note: As MCP adoption grows, this server will work with any new MCP-compatible AI assistant or development environment without modification.

Fine-tune your server's behavior:

{
  "server": {
    "name": "Claude Custom Prompts MCP Server",
    "version": "1.0.0",
    "port": 9090
  },
  "prompts": {
    "file": "promptsConfig.json",
    "registrationMode": "name"
  },
  "transports": {
    "default": "stdio",
    "sse": { "enabled": false },
    "stdio": { "enabled": true }
  }
}

Structure your AI command library:

{
  "categories": [
    {
      "id": "development",
      "name": "🔧 Development",
      "description": "Code review, debugging, and development workflows"
    },
    {
      "id": "analysis",
      "name": "📊 Analysis",
      "description": "Content analysis and research prompts"
    },
    {
      "id": "creative",
      "name": "🎨 Creative",
      "description": "Content creation and creative writing"
    }
  ],
  "imports": [
    "prompts/development/prompts.json",
    "prompts/analysis/prompts.json",
    "prompts/creative/prompts.json"
  ]
}

# Research Analysis Chain

## User Message Template

Research {{topic}} and provide {{analysis_type}} analysis.

## Chain Configuration

Steps: research → extract → analyze → summarize
Input Mapping: {topic} → {content} → {key_points} → {insights}
Output Format: Structured report with executive summary

# {{ title | title }} Analysis

## Context
{% if previous_analysis %}
Building upon previous analysis: {{ previous_analysis | summary }}
{% endif %}

## Requirements
{% for req in requirements %}
{{loop.index}}. **{{req.priority | upper}}**: {{req.description}}
   {% if req.examples %}
   Examples: {% for ex in req.examples %}{{ex}}{% if not loop.last %}, {% endif %}{% endfor %}
   {% endif %}
{% endfor %}

## Focus Areas
{% set focus_areas = focus.split(',') %}
{% for area in focus_areas %}
- {{ area | trim | title }}
{% endfor %}

# Update prompts on-the-fly
>>update_prompt id="analysis_prompt" content="new template"

# Add new sections dynamically
>>modify_prompt_section id="research" section="examples" content="new examples"

# Hot-reload everything
>>reload_prompts reason="updated templates"

// Health Check Response
{
  healthy: true,
  modules: {
    foundation: true,
    dataLoaded: true,
    modulesInitialized: true,
    serverRunning: true
  },
  performance: {
    uptime: 86400,
    memoryUsage: { rss: 45.2, heapUsed: 23.1 },
    promptsLoaded: 127,
    categoriesLoaded: 8
  }
}

We're building the future of AI prompt engineering! Join our community:

• 🐛 Found a bug? Open an issue
• 💡 Have an idea? Start a discussion
• 🔧 Want to contribute? Check our Contributing Guide
• 📖 Need help? Visit our Documentation

Released under the MIT License - see the file for details.