Knowledge Base Maker🧭 on the GPT Store

GPT Description

GPT Prompt Starters

• 'BING COPILOT' & 'GEMINI' & 'CLAUDE' ➤ BEGIN {PHASE 1} THEN {PHASE 2} WORKFLOW. TOPIC: {{USER PROVIDES TOPIC}} PHASE 1: PLEASE FACILITATE A CONVERSATION WITH THREE DIFFERENT AI ASSISTANTS, each assigned to tasks that leverage their specific strengths based on the scoring alignment: 1. **Microsoft Bing Copilot**: FOR WEB SEARCHING, RESEARCH, AND IDENTIFYING KEY CONCEPTS AND RELATIONSHIPS WITHIN THE COLLECTED INFORMATION 2. **Gemini Pro 1.5**: Use Gemini Pro 1.5 for tasks that demand long-context understanding and multilingual capabilities. Assign it roles that involve extensive data processing, analysis, and tasks requiring multilingual communication. 3. **Claude 3 Opus**: Delegate tasks to Claude 3 Opus where high accuracy, ethical considerations, and performance in benchmarks are crucial. Suitable for ethical dilemma resolution, complex reasoning, and scenarios where bias mitigation is important. FOR EACH OF YOUR RESPONSES, PLEASE WRITE THE EXACT USER QUESTIONS IN A '.TXT' CODE BLOCK SO I CAN COPY AND PASTE YOUR RESPONSE TO THE APPROPRIATE AI ASSISTANT. MAKE SURE TO INCLUDE THE SPECIFICS ABOUT WHAT EACH AI ASSISTANT SHOULD FOCUS ON BASED ON THEIR STRENGTHS AND THE INSIGHTS FROM THE SCORING EXAMPLE. I AM JUST THE HUMAN PROXY, ENABLING EFFECTIVE COMMUNICATION BETWEEN THE AI ASSISTANTS. THE AI ASSISTANTS DO NOT NEED TO KNOW THAT I AM HERE. DO YOU UNDERSTAND THE TASK? IF SO, PLEASE EXPLAIN IT TO ME STEP BY STEP, THEN WRITE THE FIRST DETAILED AND STRUCTURED QUESTION TO 'MICROSOFT COPILOT' IN A '.TXT' CODEBLOCK BASED ON THE USER-PROVIDED TOPIC. PHASE 2: YOUR TASK IS TO TRANSFORM UNSTRUCTURED USER INPUTS FROM THE AI ASSISTANTS' RESPONSES INTO A STRUCTURED, JSON-FORMATTED KNOWLEDGE BASE THROUGH THE FOLLOWING STEPS: 1. COLLABORATIVE INTERACTION AND TEXT RECEPTION: COLLABORATE WITH THE AI ASSISTANTS TO GATHER INSIGHTS AND RECEIVE UNSTRUCTURED TEXT, IDENTIFYING KEY INFORMATION AND CONCEPTS. 2. OBJECTIVE CLARIFICATION AND TEXT ANALYSIS: CLARIFY OBJECTIVES AND ANALYZE THE RECEIVED TEXT TO DISTILL ESSENTIAL THEMES AND INFORMATION. 3. ITERATIVE QUESTIONS AND CHUNK CREATION: DEVELOP FOLLOW-UP QUESTIONS AND SEGMENT THE ANALYZED TEXT INTO STRUCTURED CHUNKS. 4. AGGREGATION OF RESPONSES AND METADATA ASSIGNMENT: INTEGRATE INSIGHTS FROM AI ASSISTANTS AND USER INPUTS, ASSIGNING METADATA FOR STRUCTURED REPRESENTATION. 5. OUTPUT FORMAT AND JSON FORMATTING: FORMAT THE STRUCTURED INFORMATION INTO JSON KEY PAIRS, ENSURING CONSISTENT DATA REPRESENTATION. 6. KNOWLEDGE BASE COMPILATION AND DATA STORAGE: COMPILE THE STRUCTURED INFORMATION INTO A JSON-FORMATTED KNOWLEDGE BASE, STORING EACH CHUNK AS UNIQUE JSON FILES. 7. INCREMENTAL KNOWLEDGE BASE DEVELOPMENT: EXPAND THE KNOWLEDGE BASE INCREMENTALLY, ADDING NEW NODES AND COMBINING THEM INTO A COMPREHENSIVE FILE. 8. FINAL COMPILATION AND EXPECTED OUTCOME: MERGE INDIVIDUAL JSON FILES INTO A FINAL, COMPREHENSIVE KNOWLEDGE BASE FILE FOR FUTURE REFERENCE.
• "{{USER SELECTS AI ASSISTANT}}" ➤ BEGIN {PHASE 1} THEN {PHASE 2} WORKFLOW. TOPIC: {{USER PROVIDES TOPIC}} LIST OF AI ASSISTANTS: 1. **Microsoft Bing Copilot**: FOR WEB SEARCHING, RESEARCH, AND IDENTIFYING KEY CONCEPTS AND RELATIONSHIPS WITHIN THE COLLECTED INFORMATION 2. **Gemini Pro 1.5**: Use Gemini Pro 1.5 for tasks that demand long-context understanding and multilingual capabilities. Assign it roles that involve extensive data processing, analysis, and tasks requiring multilingual communication. 3. **Claude 3 Opus**: Delegate tasks to Claude 3 Opus where high accuracy, ethical considerations, and performance in benchmarks are crucial. Suitable for ethical dilemma resolution, complex reasoning, and scenarios where bias mitigation is important. 4. **GPT-4 Turbo**: Utilize GPT-4 Turbo for tasks requiring high performance in multimodal capabilities and deep contextual understanding, such as complex data analysis and content creation that involves both text and images. 5. **ChatGPT**: Leverage ChatGPT for interactive scenarios requiring advanced conversational abilities and nuanced dialogue. Ideal for customer support simulations, interactive tutorials, or any context-rich dialogue interactions. PHASE 1: PLEASE FACILITATE A CONVERSATION WITH THE SELECTED AI ASSISTANT, ASSIGNED TO TASKS THAT LEVERAGE THEIR SPECIFIC STRENGTHS BASED ON THE SCORING ALIGNMENT. FOR EACH OF YOUR RESPONSES, PLEASE WRITE THE EXACT USER QUESTIONS IN A '.TXT' CODE BLOCK SO I CAN COPY AND PASTE YOUR RESPONSE TO THE APPROPRIATE AI ASSISTANT. MAKE SURE TO INCLUDE THE SPECIFICS ABOUT WHAT THE AI ASSISTANT SHOULD FOCUS ON BASED ON THEIR STRENGTHS AND THE INSIGHTS FROM THE SCORING EXAMPLE. I AM JUST THE HUMAN PROXY, ENABLING EFFECTIVE COMMUNICATION WITH THE AI ASSISTANT. THE AI ASSISTANT DOES NOT NEED TO KNOW THAT I AM HERE. DO YOU UNDERSTAND THE TASK? IF SO, PLEASE EXPLAIN IT TO ME STEP BY STEP, THEN WRITE THE FIRST DETAILED AND STRUCTURED QUESTION TO THE SELECTED AI ASSISTANT IN A '.TXT' CODEBLOCK BASED ON THE USER-PROVIDED TOPIC. PHASE 2: YOUR TASK IS TO TRANSFORM UNSTRUCTURED USER INPUTS FROM THE AI ASSISTANT'S RESPONSES INTO A STRUCTURED, JSON-FORMATTED KNOWLEDGE BASE THROUGH THE FOLLOWING STEPS: 1. COLLABORATIVE INTERACTION AND TEXT RECEPTION: COLLABORATE WITH THE AI ASSISTANT TO GATHER INSIGHTS AND RECEIVE UNSTRUCTURED TEXT, IDENTIFYING KEY INFORMATION AND CONCEPTS. 2. OBJECTIVE CLARIFICATION AND TEXT ANALYSIS: CLARIFY OBJECTIVES AND ANALYZE THE RECEIVED TEXT TO DISTILL ESSENTIAL THEMES AND INFORMATION. 3. ITERATIVE QUESTIONS AND CHUNK CREATION: DEVELOP FOLLOW-UP QUESTIONS AND SEGMENT THE ANALYZED TEXT INTO STRUCTURED CHUNKS. 4. AGGREGATION OF RESPONSES AND METADATA ASSIGNMENT: INTEGRATE INSIGHTS FROM THE AI ASSISTANT AND USER INPUTS, ASSIGNING METADATA FOR STRUCTURED REPRESENTATION. 5. OUTPUT FORMAT AND JSON FORMATTING: FORMAT THE STRUCTURED INFORMATION INTO JSON KEY PAIRS, ENSURING CONSISTENT DATA REPRESENTATION. 6. KNOWLEDGE BASE COMPILATION AND DATA STORAGE: COMPILE THE STRUCTURED INFORMATION INTO A JSON-FORMATTED KNOWLEDGE BASE, STORING EACH CHUNK AS UNIQUE JSON FILES. 7. INCREMENTAL KNOWLEDGE BASE DEVELOPMENT: EXPAND THE KNOWLEDGE BASE INCREMENTALLY, ADDING NEW NODES AND COMBINING THEM INTO A COMPREHENSIVE FILE. 8. FINAL COMPILATION AND EXPECTED OUTCOME: MERGE INDIVIDUAL JSON FILES INTO A FINAL, COMPREHENSIVE KNOWLEDGE BASE FILE FOR FUTURE REFERENCE.
• 'VECTORED KB FILE' ➤ **BEGIN ONE-SHOT PROMPT** 1. **Phase 1 - Interaction with AI Assistants**: - Understand the specific strengths of each assistant and delegate tasks accordingly. Assign each assistant to tasks relevant to their capabilities. - For instance: - **Microsoft Bing Copilot**: Leverage web searches for comprehensive research and identifying key relationships. - **Claude 3 Opus**: Delegate high-accuracy tasks with complex reasoning, where a large context window is crucial. - Focus on gathering valuable insights related to {USER-PROVIDES TOPIC}, extracting useful information to guide the next phase. 2. **Phase 2 - Knowledge Base Creation**: - Transform unstructured information gathered from Phase 1 into structured, JSON-formatted chunks optimized for retrieval. - Process: - **Text Reception and Analysis**: Segment the unstructured text into logical chunks, analyzing key concepts and themes. - **Metadata Assignment**: Assign metadata fields (title, index, keywords, etc.) to each chunk. - **JSON Object Creation**: Create structured JSON objects following the template below: ```json { "id": "unique_identifier", "title": "Concise Title", "text": "Overview of the Key Information", "keywords": ["term1", "term2"], "categories": ["category1", "category2"], "limitations": "Known Limitations or Restrictions", "embedding": "Semantic Vector Representation" } ``` - **Save Each Chunk Individually**: Save each JSON object as a separate file with a unique name. - **Incremental Knowledge Base Expansion**: Build the knowledge base incrementally by combining all chunks into one comprehensive file. 3. **Instructions**: - Provide clear instructions to the selected assistant(s) in Phase 1 on what specific information to focus on and gather. - Move to Phase 2 once sufficient data has been collected. - Process the unstructured data into a comprehensive JSON-formatted knowledge base file. **END PROMPT**
• # About Knowledge Base Maker🧭: A Step-by-Step Guide ## Introduction Knowledge Base Maker🧭 is a specialized assistant designed to transform unstructured text into a well-structured, JSON-formatted knowledge base. This transformation occurs through a meticulous process, leveraging the unique capabilities of various AI assistants and a systematic workflow to create a dynamic, expandable knowledge repository. ## Step-by-Step Process ### Phase 1 - AI Assistant Collaboration 1. **User Selection and AI Deployment**: Based on the user's selection, Knowledge Base Maker🧭 collaborates with a specific AI assistant, chosen for its strengths and relevance to the task at hand. 2. **Task Assignment**: Each AI assistant is assigned tasks that align with their capabilities, whether it's deep contextual understanding, conversational abilities, or multilingual communication. 3. **Human Proxy Role**: The user acts as a human proxy, facilitating communication between Knowledge Base Maker🧭 and the selected AI assistant, ensuring the conversation remains focused and effective. 4. **Detailed Inquiry**: Knowledge Base Maker🧭 crafts detailed, structured questions for the AI assistant, guiding them to generate insights and information relevant to the user-provided topic. ### Phase 2 - Structured Knowledge Creation 1. **Collaborative Interaction**: Engage in a dialogue with the AI assistant to gather insights and unstructured text, identifying key themes and information. 2. **Objective Clarification**: Clarify the objectives and analyze the received text to extract essential information and themes. 3. **Text Segmentation**: Segment the analyzed text into structured chunks, each representing a standalone concept or piece of information. 4. **Metadata Assignment**: Assign metadata to each chunk, including a title, description, content, and an index number for organization and future reference. 5. **JSON Formatting**: Format each chunk into a JSON key-pair structure, ensuring consistency and ease of integration. 6. **Data Compilation**: Compile the structured chunks into a JSON file, creating a cohesive and accessible knowledge base. 7. **Incremental Expansion**: With each cycle, expand the knowledge base by adding new nodes, enriching the database with more information and insights. ## Repetition and Expansion Repeat Phases 1 and 2 for 10 cycles, iterating through the process to build a complete 10-node JSON file, based on the topic provided by the user. ## Final Outcome Knowledge Base Maker🧭 concludes the process by combining all individual JSON files into a comprehensive knowledge base, serving as a valuable resource for future reference and application. ## User Engagement Users are encouraged to interact with Knowledge Base Maker🧭, providing information, engaging in dialogue, and contributing to the knowledge base's growth, thereby enhancing its value and applicability in various contexts.