hemanth/paws-on-mcp

3.9

paws-on-mcp is hosted online, so all tools can be tested directly either in theInspector tabor in theOnline Client.

If you are the rightful owner of paws-on-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

Paws-on-MCP is a comprehensive Model Context Protocol (MCP) server implementing the latest MCP 2025-03-26 specification, designed to demonstrate advanced MCP capabilities.

Try paws-on-mcp with chat:

Tools

Functions exposed to the LLM to take actions

get_server_prompts

List all available prompt templates.

search_hackernews

Search HackerNews stories by title.

Args:
    query: Search term to look for in story titles
    limit: Maximum number of stories to return (default: 5, max: 20)
    
Returns:
    List of matching stories

get_github_repo_info

Get detailed information about a specific GitHub repository.

Args:
    owner: Repository owner (username or organization)
    repo: Repository name
    
Returns:
    Detailed repository information

create_sampling_request

Create a sampling request according to MCP 2025-03-26 specification.

This tool demonstrates proper MCP sampling by creating requests that clients
can process to get LLM completions with server context and enhanced model preferences.

Args:
    prompt: The prompt to send to the LLM
    context_data: Optional context data to include
    max_tokens: Maximum tokens to generate
    temperature: Sampling temperature (0.0 to 1.0)
    model_hint: Optional model name hint (e.g., "claude-3-sonnet", "gpt-4")
    intelligence_priority: How much to prioritize intelligence (0.0-1.0)
    cost_priority: How much to prioritize cost efficiency (0.0-1.0)
    speed_priority: How much to prioritize response speed (0.0-1.0)
    
Returns:
    Properly formatted MCP sampling request per 2025-03-26 spec

analyze_hackernews_trends_with_ai

Analyze HackerNews trends using AI through sampling.

This tool demonstrates how servers can use sampling to get AI analysis
of data they collect.

Args:
    topic: Topic to analyze in HackerNews stories
    count: Number of stories to analyze
    analysis_type: Type of analysis (brief, detailed, comprehensive)
    
Returns:
    AI analysis request with collected data

code_review_with_ai

Get AI-powered code review insights for a GitHub repository.

Demonstrates using sampling to get AI analysis of repository data.

Args:
    repo_owner: GitHub repository owner
    repo_name: Repository name
    review_focus: Focus area (security, performance, architecture, general)
    
Returns:
    AI code review request with repository context