mcp_prompt_mapper

mcp_prompt_mapper

Introduction

mcp_prompt_mapper is an open-source library designed to generate optimal prompts tailored for Claude, Grok, and OpenAI APIs. It transforms input from mcp_input_analyzer into structured, efficient prompts that can be used to build various parts of the MCP server, including resources, tools, and additional prompts.

Features

• Prompt Templating: Generate custom templates for creating resources and tools.
• Custom Output Formats: Supports both YAML and JSON formatted outputs optimized for Claude.
• Cross-API Compatibility: Works seamlessly with Claude, Grok, and OpenAI APIs.
• Schema-aware Auto-complete Prompts: Ensure prompts are schema-compliant using auto-completion features.
• Streaming Input Parsing: Efficiently parse streaming inputs directly in Claude Desktop.

Installation Instructions

To install mcp_prompt_mapper, you can use pip:

pip install mcp_prompt_mapper

Alternatively, if you prefer to install from the source, clone this repository and run setup.py:

git clone https://github.com/your-repo/mcp_prompt_mapper.git
cd mcp_prompt_mapper
python setup.py install

Usage Examples

Basic Example

Here is a simple example of how to use mcp_prompt_mapper with default settings.

from mcp_prompt_mapper import PromptMapper

# Initialize the PromptMapper
prompt_mapper = PromptMapper()

# Sample input from mcp_input_analyzer
input_data = {
    "resource_type": "database",
    "tool_name": "sql_query_tool"
}

# Generate prompts
prompts = prompt_mapper.generate_prompts(input_data)

print(prompts)

Advanced Example with Custom Output Format

This example demonstrates how to generate prompts in YAML format.

from mcp_prompt_mapper import PromptMapper

# Initialize the PromptMapper with YAML output format
prompt_mapper = PromptMapper(output_format='yaml')

# Sample input from mcp_input_analyzer
input_data = {
    "resource_type": "api_gateway",
    "tool_name": "http_request_tool"
}

# Generate prompts
prompts = prompt_mapper.generate_prompts(input_data)

print(prompts)

Streaming Input Example

Here's how you can handle streaming input for Claude Desktop.

from mcp_prompt_mapper import PromptMapper

# Initialize the PromptMapper with YAML output format
prompt_mapper = PromptMapper(output_format='yaml')

# Simulate streaming input
streaming_input = [
    {"resource_type": "database", "tool_name": "sql_query_tool"},
    {"resource_type": "api_gateway", "tool_name": "http_request_tool"}
]

for data in streaming_input:
    prompts = prompt_mapper.generate_prompts(data)
    print(prompts)

API Documentation

Class PromptMapper

• Initialization

  • __init__(self, output_format='json'): Initializes the PromptMapper instance. Accepts an optional output_format parameter that defaults to 'json'. Supported values are 'json' and 'yaml'.

• Method generate_prompts

  • generate_prompts(self, input_data): Takes a dictionary of input data and generates prompts based on the provided schema. Returns the generated prompts in the specified output format.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Ensure you replace `"https://github.com/your-repo/mcp_prompt_mapper.git"` with the actual URL of your repository if it's different. Additionally, ensure that any other paths or references are correctly updated to match your project setup.


## ⚠️ Development Status

This library is currently in early development. Some tests may be failing with the following issues:

Contributions to fix these issues are welcome! Please submit a pull request if you have a solution.