ArXiv MCP Server

"""Prompt definitions for arXiv MCP server with research journey support.""" from mcp.types import ( Prompt, PromptArgument, ) # Define all prompts PROMPTS = { "research-discovery": Prompt( name="research-discovery", description="Begin research exploration on a specific topic", arguments=[ PromptArgument( name="topic", description="Research topic or question", required=True ), PromptArgument( name="expertise_level", description="User's familiarity (beginner/intermediate/expert)", required=False, ), PromptArgument( name="time_period", description="Time period for search (e.g., '2023-present')", required=False, ), PromptArgument( name="domain", description="Academic domain (e.g., computer_science/physics/biology)", required=False, ), ], ), "deep-paper-analysis": Prompt( name="deep-paper-analysis", description="Analyze a specific paper in detail", arguments=[ PromptArgument( name="paper_id", description="arXiv paper ID", required=True ), ], ), "literature-synthesis": Prompt( name="literature-synthesis", description="Synthesize findings across multiple papers", arguments=[ PromptArgument( name="paper_ids", description="Comma-separated list of arXiv paper IDs", required=True, ), PromptArgument( name="synthesis_type", description="Synthesis type (themes/methods/timeline/gaps/comprehensive)", required=False, ), PromptArgument( name="domain", description="Academic domain (e.g., computer_science/physics/biology)", required=False, ), ], ), "research-question": Prompt( name="research-question", description="Formulate research questions based on literature", arguments=[ PromptArgument( name="paper_ids", description="Comma-separated list of arXiv paper IDs", required=True, ), PromptArgument( name="topic", description="Research topic or question", required=True ), PromptArgument( name="domain", description="Academic domain (e.g., computer_science/physics/biology)", required=False, ), ], ), }