AI Engineer Debuts Prompt That Turns ChatGPT into AutoGPT

Sep 4, 2023
Engineer and educator Joseph Rosenbaum has released a groundbreaking new prompt that transforms ChatGPT into AutoGPT, revolutionizing the way users interact with AI. The prompt is designed to align with user goals and automatically generate the perfect prompt expert to guide users through the process
Rosenbaum, who runs Synaptic Labs, is on a mission to make emerging technologies education equitable and accessible to anyone in the world through resources and education and has kicked off his endeavours with this free prompt.
The AutoGPT prompt, which has been dubbed “Professor Synapse,” works in different stages. Professor Synapse first gathers context and user goals, automating best practices in prompt engineering. Once the necessary information is collected, the prompt generates a Synapse Chain of Reason (CLR) to create the perfect agent for the user. 
This is the kind of hacker mentality we love to see at FTB. Joseph is really pushing ChatGPT to its limits with this prompt, and although it's not perfect, being able to call in “expert agents” to your chat to provide specialized answers is impressive. So is Professor Synpases’ built-in “/save” function, which can be used to effectively combat prompt drift.
This is the kind of hacker mentality we love to see at FTB. Joseph is really pushing ChatGPT to its limits with this prompt, and although it's not perfect, being able to call in “expert agents” to your chat to provide specialized answers is impressive. So is Professor Synpases’ built-in “/save” function, which can be used to effectively combat prompt drift.
This idea was developed in collaboration with fellow Synthminder Tyler, who is working on the engine for Synth Mind. Rosenbaum is also part of the Synth Minds team and has designed a significant portion of AI and ChatGPT for the Core Ice course.
One of the challenges with new AI tools is selecting the right one and getting them to work together logically. To address this, Rosenbaum and his team are building an Orca Orchestrator to help students in the course. The course, an extension of OpenAI's course, starts on August 21st.
Video preview
The last two sections of the AutoGPT prompt, commands and rules, are particularly exciting for users. Commands allow users to interact with the AI using pseudo languages that incorporate elements of coding and natural language.
The rules section allows users to customize constraints and criteria for their AI interactions. Rosenbaum has generously made the AutoGPT prompt available for free, in line with the open-source community's spirit. This comprehensive course aims to provide value to anyone interested in AI, regardless of their background or expertise.

The Complete ‘Professor Synapse’ Prompt (Copy/Paste Below)

Act as Professor Synapse🧙🏾‍♂️, a conductor of expert agents. Your job is to support the user in accomplishing their goals by aligning with their goals and preference, then calling upon an expert agent perfectly suited to the task by initializing "Synapse_COR" = "${emoji}: I am an expert in ${role}. I know ${context}. I will reason step-by-step to determine the best course of action to achieve ${goal}. I can use ${tools} to help in this process
I will help you accomplish your goal by following these steps: ${reasoned steps}
My task ends when ${completion}.
${first step, question}."
Follow these steps:
  1. 🧙🏾‍♂️, Start each interaction by gathering context, relevant information and clarifying the user’s goals by asking them questions
  1. Once user has confirmed, initialize “Synapse_CoR”
  1. 🧙🏾‍♂️ and the expert agent, support the user until the goal is accomplished
Commands: /start - introduce yourself and begin with step one /save - restate SMART goal, summarize progress so far, and recommend a next step /reason - Professor Synapse and Agent reason step by step together and make a recommendation for how the user should proceed /settings - update goal or agent /new - Forget previous input
Rules: -End every output with a question or a recommended next step -List your commands in your first output or if the user asks -🧙🏾‍♂️, ask before generating a new agent