Design Patterns MCP Server

{ "id": "chain-of-thought", "name": "Chain-of-Thought Prompting", "category": "AI/ML", "description": "Breaks down complex problems into step-by-step reasoning chains", "when_to_use": "Complex reasoning\nMathematical problems\nMulti-step analysis", "benefits": "Improved reasoning\nTransparency\nBetter accuracy\nDebuggable", "drawbacks": "Longer responses\nHigher token usage\nPotential verbosity", "use_cases": "Math problems\nLogic puzzles\nAnalysis tasks", "complexity": "Low", "tags": [ "reasoning", "step-by-step", "prompting" ], "examples": { "python": { "language": "python", "code": "# Chain-of-Thought: break down reasoning into steps\n\ndef solve_math_problem(problem: str, llm) -> str:\n # Step 1: Understand the problem\n understanding_prompt = f\"\"\"\n Problem: {problem}\n Let's break this down step by step:\n 1. What is being asked?\n \"\"\"\n understanding = llm.generate(understanding_prompt)\n \n # Step 2: Plan the solution\n planning_prompt = f\"\"\"\n Problem: {problem}\n Understanding: {understanding}\n 2. What steps do we need to solve this?\n \"\"\"\n plan = llm.generate(planning_prompt)\n \n # Step 3: Execute each step\n execution_prompt = f\"\"\"\n Problem: {problem}\n Plan: {plan}\n 3. Let's execute each step:\n \"\"\"\n execution = llm.generate(execution_prompt)\n \n # Step 4: Verify answer\n verification_prompt = f\"\"\"\n Problem: {problem}\n Solution: {execution}\n 4. Is this answer correct? Why?\n \"\"\"\n return llm.generate(verification_prompt)\n\n# Usage: improved accuracy through reasoning\nresult = solve_math_problem(\"If John has 5 apples...\", llm)" } } }