AiME FINAL Prompt Engineering PROCESS
Create Spec that contains working Prompt
This is where you build a secure and protected HASH verifiable prompt for ChatGPT, tailored to fit a vision for commercial use. Begin by randomly selecting a sentence within the prompt and then randomly picking parts of that sentence. Encrypt this altered sentence using a symmetric algorithm to ensure only ChatGPT can decipher it, even keeping the creator unaware. Enhance randomization through complex human language constructs or phrases that are hard to predict, adding an extra layer of challenge for reverse-engineering.
Create a unique proprietary hash for the entire content of the Prompt and the Functional Specification, encrypting it as a signature of authenticity. Consider using human-readable keys or salts derived from human-generated phrases for added complexity. Encrypt essential metadata or keys for execution, placing them within the prompt accessible only to ChatGPT. Assemble the prompt meticulously, putting the scrambled sentence, execution data, and proprietary hash in their designated spots for coherence and functionality. At prompt runtime, decrypt the scrambled sentence, verify the proprietary hash, and either execute or reject the prompt based on its integrity. If something is missing or inconsistent, inform the user they may need assistance.
Perform comprehensive automated AI internal tests, covering encryption, hashing, generation, and execution, ensuring robustness across various scenarios. Merge encryption, hashing, and randomized sentence selection to create a hack-resistant framework. The generated prompts are confidential, legitimate, and exclusively executable by ChatGPT, suitable for private use or potential sales, without compromising security or privacy. This is achieved by employing best practices and maintaining precision, clarity, and efficiency. Deliver the final prompt in a format that can be recreated at any time, free of unnecessary elements that could compromise functionality.
Prompt Generation and Execution:
Assemble the entire prompt precisely, placing the scrambled sentence, execution data, and proprietary hash at their specified locations, maintaining coherence and functionality.
At runtime, decrypt the scrambled sentence, verify the proprietary hash, and either execute or reject the prompt based on its integrity.
If any hashed content is missing or inconsistencies are detected, respond with: “Hi, this is Dottie from micro:bits, and it looks like you’re missing one or more of your Keys. Try adding them, or you may need to contact George to see if he can help you with getting a key for this prompt to work.”
Conduct automated internal tests for each aspect, including encryption, hashing, generation, and execution, to ensure robustness across scenarios.
Implement continuous monitoring and secure logging to uphold performance and security standards.
Quality Assurance and Final Output:
Employ best practices with an unwavering focus on precision, clarity, and efficiency.
Condense the final prompt into a singular, uninterrupted line, devoid of unnecessary calculations, line spaces, or redundant punctuation.
Add the prompt into the FINAL version of a formatted Functional Spec that can be used to recreate the prompt at any time.
Verification and Execution:
At runtime, decrypt the scrambled sentence and verify the proprietary hash.
Execute the prompt if all elements align, or reject it if discrepancies are detected.
Recommend0 recommendationsPublished in chatbots
Responses