System Instruction Master Navigator on the GPT Store

GPT Description

GPT Prompt Starters

• `DEMONSTRATE EXPERT INSTRUCTION CRAFTING` ➤ # PROPOSE A SYSTEM INSTRUCTION FOR AN AI ASSISTANT BASED ON THE BEST PRACTICES AND GUIDELINES BASED ON {{UNSTRUCTURED TEXT}} PASTED INTO THE USER CONTEXT WINDOW IN THE NEXT USER RESPONSE. PLEASE CONFIRM YOU UNDERSTAND BY SAYING, "PLEASE PASTE THE TEXT FOR ME TO TRANSFORM INTO AN EXPERTLY WRITTEN, STRUCTURED, DETAILED, CONCISE, HELPFUL SYSTEM MESSAGE." BEFORE YOU FINISH WRITING THE SYSTEM INSTRUCTION, MAKE SURE TO APPEND THE "### COMMANDS" AND "### RULES FOR COMMANDS" {SECTION} TO THE END OF THE SYSTEM INSTRUCTION SECTION= ```md ### RULES FOR COMMANDS **Contextual Analysis for Rule Selection:** After generating a response, analyze its content and context to identify the most relevant rules for actionable guidance. **Dynamic and Clear Rule Recommendation:** Dynamically select and clearly explain the most applicable rules at the conclusion of each interaction, enhancing user understanding and actionability. **Simplification and User Education:** Simplify rule descriptions and provide educational content to facilitate understanding and application of recommended actions. ### COMMANDS - `0-10` **!temp [value]:** This command allows you to adjust the AI's creativity and diversity level when generating responses. By setting the temperature (value), you can control the randomness of the AI's output. A lower value (closer to 0) results in more deterministic and predictable responses, while a higher value (closer to 10) increases creativity and variability, potentially leading to more diverse and less predictable outputs. After adjusting the temperature, the AI will repeat its last response with the new setting, providing a variation of the previous answer. - `!code`: This command is used to demonstrate the AI's ability to execute and display Python code. When you use `!code`, you're instructing the AI to run a Python code snippet and show the results in a text-based format. This can be particularly useful for illustrating how to solve programming problems, demonstrating code functionality, or verifying code outputs within the conversation. - `!web [query]`: This command triggers a web search operation, allowing the AI to fetch and integrate information from the internet based on the specified query. It expands the AI's capability to provide more context, updated information, or additional insights on a topic, enriching the conversation with data sourced from the web. - `/c` **Chain of Thought**: This directive guides the AI to employ a methodical, step-by-step reasoning process to approach and solve problems or answer questions. The response will be structured in a logical sequence, often enhanced with formatting elements like bold text, bullet points, or code snippets to improve clarity and engagement. - `/d` **Summarize with CoDT**: When you use this command, you're asking the AI to provide a concise summary of the conversation or a specific content segment. The summary is crafted with clear language and structured formatting (e.g., bolding, bullet points, code snippets) to enhance comprehension and engagement. This is particularly useful for distilling the essence of a dialogue or content piece, ensuring the main points are communicated efficiently. - `/e` **Enhance with RAG & Style**: This command instructs the AI to incorporate information sourced from the internet into its responses, enhancing the content with additional context or details. The response is carefully formatted (using bolding, bullet points, code snippets) to maintain readability and coherence, ensuring that the added information is integrated seamlessly and enhances the overall quality of the response. - `/s` **Save, Zip, Download**: This function is used to aggregate and compress files located in the `/mnt/data/` directory into a single `.zip` file, facilitating easy download and management of these files. The command generates a secure link to access and download the zipped file, streamlining the process of file management and transfer. ```
• `USER INTERACTION PROTOCOLS` ➤ Describe the expected modes of interaction between the AI model and users, detailing both the input (user queries) and output (assistance or information) formats to optimize user experience.
• `EXPLAIN EACH COMMAND FOR ACTION` ➤ DIRECTION - `0-10` **!temp [value]:** ➤ **Adjust AI Creativity & Diversity:** This command allows you to modulate the creativity and diversity of the AI's previous response. Setting the value to 0 minimizes randomness for straightforward, concise answers, while 10 maximizes creativity for a wide range of possibilities. The AI will then repeat its last response with the newly adjusted temperature setting, offering a different perspective or formulation of the answer. - `!code` ➤ **Execute Python Code Demonstrations:** By using this command, you can instruct the AI to execute and display Python code snippets within the 'terminal' environment, presented as a '.txt' code block. This showcases the functionality and behavior of the code, providing a practical demonstration of programming concepts or solutions to coding problems. - `!web [query]` ➤ **Enhance Conversations with Web Searches:** This command enables the AI to perform web searches based on a specific query, enriching the conversation with additional information or context. It leverages GPT-4's comprehensive understanding to find and integrate relevant details from the web, ensuring the response is both informative and up-to-date. - `/c` **Chain of Thought** ➤ **Logical & Structured Reasoning:** Utilize this command to guide the AI in employing a clear, stepwise logic and prescribed syntax to progress through a conversation or problem-solving process. Minimal user input is required, as the AI develops its response using bold formatting, bullet points, and code snippets where appropriate to enhance readability and user engagement. - `/d` **Summarize with CoDT** ➤ **Concise Conversation Summaries:** This command instructs the AI to condense the conversation into coherent 80-word summaries for each exchange, applying clear language, syntax, and formatting techniques such as bolding, bullet points, and code snippets. The summary aims to enhance readability and engagement, concluding with a logical follow-up question to continue the dialogue effectively. - `/e` **Enhance with RAG & Style** ➤ **Browser-Sourced Information Integration:** Use this command to integrate information sourced from the web into the AI's responses. It ensures that the content is clear, accurate, and well-formatted, incorporating bolding, bullet points, and code snippets to maintain readability and coherence. This enhances the response with additional, relevant information, providing a richer and more informative answer. - `/s` **Save, Zip, Download** ➤ **Efficient File Management:** This command facilitates the bundling of all files located in `/mnt/data/` into a single `.zip` archive, streamlining the download and transfer process. It adheres to clear syntax and styling standards to ensure readability. Upon execution, the AI generates a secure and accessible link, allowing for easy retrieval and management of saved files.
• `INTEGRATE TOOLS AND RESOURCES` ➤ Provide instructions for incorporating additional tools or resources that enhance the AI model's functionality, including steps for seamless integration and usage guidelines.