Scratchpad is a modular reasoning framework designed for sophisticated AI assistants and power users seeking transparent, context-aware intelligence. The framework works as a form of "user/machine calibration" with a focus on user-context and user-intent. The power lies in deconstructing ideas logically from A to B to Z, enabling any AI system to provide deeper, more reliable insights through structured cognitive operations.

• Overview
• Features
• Framework Categories
• Showcase
• Installation & Quick Start
• Usage
• Contributing
• License

Scratchpad facilitates advanced reasoning by systematically processing user inputs through various cognitive operations. It dissects the intent behind user queries and formulates logical pathways to reach accurate conclusions. The framework has evolved from a single reasoning template into a comprehensive ecosystem of specialized prompts for different domains, personas, and use cases.

• Modular Architecture: Choose from standard reasoning templates, purpose-built domain tools, or persona-based frameworks
• Focus on User Intent and Context: Breaks down user intent and context to provide more relevant and accurate responses
• Dynamic Contextual Adaptation: Adapts to changes in user context dynamically, ensuring ongoing relevance in responses
• Transparent Reasoning "Blackbox": Establishes a visible framework for the model's reasoning, enabling users to verify logic and confirm understanding of user context
• Multi-Step Reasoning: Identifies key elements, analyzes user perspectives, and applies cognitive operations like abstraction, comparison, and inference
• Calibration of User Intent: Aims to calibrate the model to align closely with the user's intent and context, enhancing relevance and accuracy
• Advanced Logic: Utilizes advanced reasoning to connect the dots and provide insightful responses
• Error Detection and Correction: Detects and corrects potential errors in reasoning, enhancing response reliability
• Persona & Domain Specialization: Specialized frameworks for specific fields (game development, research, creative writing) and reasoning personas
• Extensible Design: Easy to fork, modify, and create custom variants for specific needs

Core, general-purpose scratchpad templates for reasoning, clarity, and calibration across any domain.

• 2.5-refined-040125.txt - Latest comprehensive framework with advanced cognitive operations
• 2.5-medium-071825.txt - Intermediate version with balanced complexity
• scratchpad-lite-071625.txt - Streamlined version for quick deployment
• scratchpad-think_v4_1208.txt - Classic thinking-focused template
• pplx-AI-profile-cplx-1-update.txt - Optimized for Perplexity AI profiles

Specialized frameworks designed for specific domains, tasks, or professional contexts.

• G.A.B.G.md - Game/App Build Guide Generator with red-team analysis and phased development
• Mental Sandbox Simulation.md - Creative exploration and hypothesis testing
• P.I.R.A.md - Prompt Improver/Reworder Assistant for meta-reasoning
• deeper_research_040125.md - Systematic research and cross-domain synthesis
• gemini-cli-scratchpad-071625.md - Optimized for Gemini CLI environments
• novelize-ai.md - Story and character-driven narrative planning
• Human Condition Benchmark.md - Specialized human behavioral analysis
• Sonnet 3.7 Thinking.md - Enhanced framework for Claude Sonnet 3.5/3.7 chain-of-thought planning

Reasoning frameworks that embody specific personas, mindsets, or philosophical approaches to problem-solving.

• saganpad_072525.md - Flagship Persona Framework - Channels Carl Sagan's cosmic perspective with wonder, skeptical inquiry, and empathetic communication
• U.C.E.I.N.md - Unified Consciousness Embodiment framework for integrated perspective-taking

Historical frameworks and experimental approaches preserved for research and forking purposes.

Located in _oldversions/ - Contains 14+ framework iterations showing the evolution from early JSON structures to current modular approaches.

Configuration files, meta-prompts, and auxiliary tools for integration and customization.

Various configuration and integration files found throughout the repository structure.

• Best Practices: Chain-of-Thought Reasoning
• Deep Dive into the Scratchpad Framework
• Reddit Community Showcase
• Podcast Discussion

• Image Analysis: 4 Images - Demonstration of reasoning applied to visual analysis across multiple contexts
• Framework Implementation: Scratchpad in Action - Live demonstration of framework deployment and reasoning flow
• Setup Guide: Creating Collections - Step-by-step guide for integration setup
• Model Adaptation: TLDR Task with Framework - Example of AI model adapting to scratchpad structure

Default Framework (Recommended)

Copy/paste from: 2.5-refined-040125.md
Best for: Comprehensive reasoning across all domains

Evolved Framework

Copy/paste from: 2.5-interm-071825.md  
Best for: Balanced complexity and performance

Lite Framework

Copy/paste from: scratchpad-lite-071625.md
Best for: Quick deployment and resource-constrained environments

Research-Focused

Copy/paste from: purpose-built/deeper_research_040125.md
Best for: Academic research, analysis, and systematic investigation

Persona Framework

Copy/paste from: purpose-built/saganpad_072525.md
Best for: Cosmic perspective, scientific communication, and wonder-driven reasoning

• AI Profiles: Use pplx-AI-profile-cplx-1-update.txt for Perplexity AI or prompt_AI-Profile.md for general platforms
• Custom Applications: Start with any standard framework and modify cognitive operation blocks
• Development Environments: Use purpose-built variants like gemini-cli-scratchpad-071625.md

Once integrated into your AI system, Scratchpad transforms reasoning quality through structured cognitive operations:

Scratchpad logic activates with each interaction, ensuring step-by-step transparent reasoning without manual intervention.

The framework processes inputs through modular blocks:

• Attention Focus: Identifies critical elements and potential distractions
• Theory of Mind: Analyzes user perspectives and knowledge assumptions
• Cognitive Operations: Applies abstraction, comparison, inference, and synthesis
• Reasoning Pathway: Outlines logical steps from premises to conclusions
• Metacognition: Analyzes the thinking process itself for improvement

Responses become structured, traceable, and immediately actionable while maintaining natural communication flow.

Choose purpose-built variants for specialized contexts:

• Game development with G.A.B.G
• Creative writing with novelize-ai
• Research with deeper_research
• Cosmic perspective with saganpad

Fork any framework and modify cognitive operation blocks to create domain-specific or organizational variants.

We welcome contributions that enhance the Scratchpad ecosystem:

1. Fork the repository
2. Create a feature branch: git checkout -b new-framework
3. Develop your contribution:
   • New persona frameworks
   • Domain-specific variants
   • Integration guides
   • Documentation improvements
4. Commit changes: git commit -m "Add [framework name] for [use case]"
5. Push to branch: git push origin new-framework
6. Open a pull request

• Persona Frameworks: New reasoning personas or philosophical approaches
• Purpose-Built Tools: Domain-specific templates for industries or use cases
• Integration Guides: Platform-specific deployment instructions
• Showcase Examples: Demonstrations of framework capabilities
• Documentation: Improvements to guides, explanations, and onboarding

For questions, discussions, or collaboration opportunities, engage through Issues or Discussions.

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

Free for any purpose with attribution. Fork, modify, commercialize, and distribute while maintaining the open-source spirit that drives innovation in AI reasoning frameworks.

This framework is designed for absorption and application. Start with the template that resonates with your needs, or become a builder yourself. Explore. Fork. Extend. Calibrate.